FEMS Microbiology Reviews, fuz016, 43, 2019, 548–575

doi: 10.1093/femsre/fuz016
Advance Access Publication Date: 10 June 2019
Review article

REVIEW ARTICLE

Cell wall peptidoglycan in Mycobacterium tuberculosis:

An Achilles’ heel for the TB-causing pathogen
Arundhati Maitra1,† , Tulika Munshi1,‡,† , Jess Healy2 , Liam T. Martin1 ,
Waldemar Vollmer3 , Nicholas H. Keep1 and Sanjib Bhakta1, *
1
Mycobacteria Research Laboratory, Institute of Structural and Molecular Biology, Department of Biological
Sciences, Birkbeck, University of London, Malet Street, London WC1E 7HX, UK, 2 Department of
Pharmaceutical and Biological Chemistry, UCL School of Pharmacy, 29–39 Brunswick Square, London WC1N
1AX, UK and 3 The Centre for Bacterial Cell Biology, Institute for Cell and Molecular Biosciences, Newcastle
University, Richardson Road, Newcastle upon Tyne, NE2 4AX, UK
∗
Corresponding author: Mycobacteria Research Laboratory, Institute of Structural and Molecular Biology, Department of Biological Sciences, Birkbeck,
University of London, Malet Street, London WC1E 7HX, UK. Tel: +44-0207631-6355; Fax: +44-0207631-6246; E-mail: sanjib.bhakta@ucl.ac.uk
One sentence summary: A comprehensive, investigative review that highlights key enzymes involved in mycobacterial cell wall peptidoglycan
metabolism as novel anti-infective drug targets.
†
Equal contribution
‡
Current address: Infection & Immunity, St. George’s, University of London, Cranmer Terrace, London SW17 0RE, UK
Editor: Chris Whitfield

ABSTRACT
Tuberculosis (TB), caused by the intracellular pathogen Mycobacterium tuberculosis, remains one of the leading causes of
mortality across the world. There is an urgent requirement to build a robust arsenal of effective antimicrobials, targeting
novel molecular mechanisms to overcome the challenges posed by the increase of antibiotic resistance in TB. Mycobacterium
tuberculosis has a unique cell envelope structure and composition, containing a peptidoglycan layer that is essential for
maintaining cellular integrity and for virulence. The enzymes involved in the biosynthesis, degradation, remodelling and
recycling of peptidoglycan have resurfaced as attractive targets for anti-infective drug discovery. Here, we review the
importance of peptidoglycan, including the structure, function and regulation of key enzymes involved in its metabolism.
We also discuss known inhibitors of ATP-dependent Mur ligases, and discuss the potential for the development of
pan-enzyme inhibitors targeting multiple Mur ligases.

Keywords: Tuberculosis (TB); cell envelope; peptidoglycan; metabolism; drug-target validation; antibiotic resistance;
antibacterial

INTRODUCTION million deaths reported due to TB in 2017 alone (WHO 2018).

An estimated 457 000 cases reported in 2017 presently harbour
Tuberculosis (TB) is a leading cause of mortality in the world
multi-drug resistant TB (MDR-TB), 8.5% of which are expected to
today and is caused by the bacterial pathogen Mycobacterium
be extensively drug resistant TB (XDR-TB) (WHO 2018), reflect-
tuberculosis. Despite innovations in diagnostics and improved
ing the urgent need to design and develop novel drugs to
access to health care, the global burden of TB remains sub-
treat TB.
stantial with around 10 million new cases of infection and 1.6

Received: 7 March 2019; Accepted: 7 June 2019

C The Author(s) 2019. Published by Oxford University Press on behalf of FEMS. This is an Open Access article distributed under the terms of the

Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and
reproduction in any medium, provided the original work is properly cited.

548

The success of M. tuberculosis as a pathogen and its innate
resistance to many antimicrobial drugs can be attributed in part
to its unique cell wall structure, which has low permeability for
many drugs and possesses a large number of efflux pumps (Jar-
lier and Nikaido 1994, Brennan and Nikaido 1995). The cell wall
is a defining characteristic of all bacteria. Amongst the many
purposes it serves, maintaining the cell-shape and withstanding
turgor are key. The varying properties of the bacterial cell wall,
especially the thickness of the peptidoglycan (PG), impart differ-
ent stain-retention properties to the bacterial cell and enable us
to categorise most bacteria into Gram-positive, Gram-negative
and acid-fast. The presence of PG across nearly all bacteria indi-
cates that it was likely to have been present in their last common
ancestor (Errington 2013). Importantly, PG is essential for bacte-
rial cell survival in most environments, thus making it a good
target for anti-infective therapy.
Mycobacteria belong to the diverse family of Actinobac-
teria. The main components of the mycobacterial cell wall
are the PG layer, mycolic acid (MA) and arabinogalactan (AG).
The mycobacterial cell wall resembles both the Gram-positive
and Gram-negative cell envelope by having a PG layer nearly
as thick as the former and an outer, waxy layer mimicking
the outer membrane of the latter (Fig. 1A). The cell wall
of mycobacteria plays a key role in intrinsic antibiotic resis-
tance and virulence (Forrellad et al. 2013; Becker and Sander
2016) but it is unclear why and how its complex structure
evolved. Vincent et al. recently proposed that the ‘mycomem-
brane’ evolved by successive horizontal acquisition of genes;
explaining the narrow distribution of the AG and MA biosyn-
thetic genes in some lineages of the Actinobacteria that evolved
later, while the PG genes are broadly distributed (Vincent et al.
2018).
PG is a dynamic structure but the factors that regulate its
composition and biogenesis in the slow-growing intra-cellular
pathogen M. tuberculosis are not well understood. The mycobac-
terial PG plays a key role in the cell’s growth, cell–cell communi-
cation and in the initiation of the host immune response. The
cell envelope of some model bacteria such as Escherichia coli,
have long been the focus of extensive research; however, these
organisms do not serve as appropriate models when studying
the unique physiology and biochemistry of M. tuberculosis. Here,
we review the different stages of PG metabolism in M. tubercu-
losis, its importance for survival of the bacilli and the potential
therapeutic targets it offers for the design of new anti-infective
drugs.
Modifications of cell wall PG in M. tuberculosis
Lying outside the cytoplasmic membrane of M. tuberculosis, the
PG layer is covalently attached to AG which itself serves as
an attachment site for unique MAs (Fig. 1A). This is referred
to as the cell wall core, or as the MA-AG-PG complex (MAPc)
(Brennan 2003). Beyond and interspersed within the MAPc
are free lipids, phosphatidyl inositol mannosides (PIM), lipo-
mannans (LM) and lipoarabinomannans (LAM). As observed
in other microorganisms (Jankute et al. 2015), the disruption
of PG synthesis or digestion of the PG layer leads to the for-
mation of L-form bacilli that display a variety of cell shapes
and morphology in mycobacteria (Udou, Ogawa and Mizuguchi
1982).
PG is made of glycan strands of alternating, β(1→4)-
linked N-acetylglucosamine (GlcNAc) and N-acetylmuramic
Maitra et al. 549
acid (MurNAc) residues. Adjacent glycan strands are con-
nected by short peptide stems with the sequence l-alanyl-
γ -d-isoglutamyl-meso-diaminopimelate-d-alanyl-d-alanine and
linked to the d-lactoyl residue of each MurNAc. Peptides protrud-
ing from adjacent glycan strands can be cross-linked between
the carboxyl group of d-Ala at the fourth position and the ε-
amino group of the meso-diaminopimelate (mDAP) in the third
position.
While the basic structure of PG is conserved, it is important
to note that the PG in mycobacteria undergoes several modifi-
cations (Fig. 1B). PG isolated from M. tuberculosis and M. smeg-
matis contains both N-acetylmuramic acid (MurNAc) and N-
glycolylmuramic acid (MurNGlyc), whereas most other bacte-
ria, including M. leprae, exclusively contain MurNAc (Mahap-
atra et al. 2005a, Petit et al. 1969; Mahapatra et al. 2008). The
presence of MurNGlyc increases the resistance of mycobacterial
PG to lysozyme (Raymond et al. 2005). Modifications such as N-
deacetylation or O-acetylation of GlcNAc and MurNAc respec-
tively, as observed in other pathogenic bacteria (Vollmer 2008)
have not been reported in mycobacteria. Consequently, the dele-
tion of the hydroxylase responsible for N-glycolylation, namH,
increases the sensitivity towards β-lactams and lysozyme in M.
smegmatis (Raymond et al. 2005). Glycolylated PG fragments are
more potent at inducing NOD2-mediated host responses such
as production of tumour necrosis factor α (Hansen et al. 2014),
which may appear to be counter-productive for the success of
an intracellular pathogen. However, the enhanced stimulation of
the host might enable M. tuberculosis to recruit phagocytic cells,
its ecological niche within the host and facilitate transmission
to a new host.
The peptide stems in PG undergo modifications such as ami-
dation of the α-carboxylic group of d-isoglutamate (d-iGlu) and
the ε-carboxylic group of the mDAP residues (Kotani et al. 1970;
Mahapatra et al. 2008). d-iGlu amidation is required for the
functioning of PG transpeptidases that cross-link the peptides
to form a robust PG (Figueiredo et al. 2012; Zapun et al. 2013)
whereas mDAP amidation is involved in modulating PG hydrol-
ysis thereby maintaining the integrity of the cell wall (Nakel,
Ghuysen and Kandler 1971; Dajkovic et al. 2017). These modi-
fications reduce the net negative charge of the cell wall, which
impairs the efficacy of antimicrobial factors such as lysozyme,
antimicrobial peptides and cell wall acting antibiotics (Girardin
et al. 2003b; Roychowdhury, Wolfert and Boons 2005). In the PG
of M. leprae, Gly replaces l-Ala in position 1 of the peptide stem,
and an additional Gly residue is attached to mDAP (Mahapatra
et al. 2008).
The PG of M. tuberculosis contains a particularly high percent-
age (70%–80%) of cross-linked peptides (Wietzerbin et al. 1974),
compared to 40%–50% in E. coli (Glauner, Holtje and Schwarz
1988). One-third of the peptide cross-links are DD-(or 4→3)
bonds between d-Ala and mDAP, and about two-thirds are LD-(or
3→3) bonds linking two mDAP residues (Wietzerbin et al. 1974;
Hett and Rubin 2008) (Fig. 1B). Although LD-crosslinks are rela-
tively rare in many other bacterial species, recent reports have
estimated them to account for 80% of the peptide cross-links
in the PG of stationary phase or dormant cells of M. tuberculo-
sis (Lavollay et al. 2008). The LD-cross-links are also enhanced to
15%–20% of all cross-links in the PG of stationary phase cells of E.
coli, but are significantly reduced (5% of all cross-links) in expo-
nentially growing cells (Pisabarro, de Pedro and Vazquez 1985;
Glauner, Holtje and Schwarz 1988; Templin, Ursinus and Holtje
1999).
Interestingly, the pattern of PG incorporation into the cell
wall of M. tuberculosis is different compared to the PG of other
 550 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5

Figure 1. (A), Differences in the cell envelope architecture of Gram-positive, acid-fast and Gram-negative bacteria. AG, arabinogalactan; GL, glycolipid; LAM, lipoarabino-
mannan; LP, lipoprotein; LPS, lipopolysaccharide; LTA, lipoteichoic acid; MA, mycolic acid; MAP, membrane-associated protein; OM, outer membrane; PG, peptidoglycan;
PM, plasma membrane; TA, teichoic acid . (B), PG structure revealing the unique features present in mycobacterial cells highlighted in blue. The stars depict residues
that undergo amidation.

rod-shaped bacteria (Daniel and Errington 2003; Hett and Rubin
2008). Most rod-shaped bacilli such as E. coli and Bacillus subtilis
elongate by inserting nascent PG along the lateral sides of the
cell (den Blaauwen et al. 2008). An actin-like protein, MreB, posi-
tions the PG-elongation machinery along the short axis of the
cell so that glycan strands are inserted circumferentially thus
maintaining the rod-shape (Domı́nguez-Escobar et al. 2011). In
contrast, mycobacteria lack MreB and other proteins important
for lateral wall elongation and instead incorporate nascent PG at
both poles of the cell (Thanky, Young and Robertson 2007; Kieser
and Rubin 2014). Mycobacteria exhibit differential incorporation
of PG at the two cell poles (old versus new cell pole) at certain
stages of the cell cycle (before/after cytokinesis); this allows for
asymmetric cell division and results in a heterogeneous popu-
lation of cells (Kieser and Rubin 2014; Botella et al. 2017b). Bara-
nowski et al. reported that the ability of the mycobacterial cell to
generate LD-cross-links (discussed later in this review) is essen-
tial in maintaining the rod-shape (Baranowski et al. 2018). How-
ever, whether this serves as the sole mechanism regulating cell
shape is yet to be confirmed.
Imaging PG
The recent application of fluorescence and atomic force
microscopy techniques together with new probes for labelling
have significantly advanced our understanding of the dynamics
of PG synthesis (Radkov et al. 2018). Localisation studies using
GFP-labelled PG biosynthetic proteins (Joyce et al. 2012; Grze-
gorzewicz et al. 2016), fluorescent d-amino acids (Siegrist et al.
2013; Meniche et al. 2014; Boutte et al. 2016; Schubert et al. 2017;
Botella et al. 2017b; Rodriguez-Rivera et al. 2018) and cell wall
targeting antibiotics (Thanky, Young and Robertson 2007; Kas-
trinsky and Barry 2010) have further emphasized the impor-
tance of PG remodelling in cell elongation, septum formation
and cell division. Aldridge et al. used microfluidics-based live-
cell imaging technique to show that the heterogeneous daugh-
ter cell populations differ in their susceptibility to antibiotics
(Aldridge et al. 2012). Botella et al. compared the spatiotempo-
ral dynamics of PG synthesis in M. smegmatis and M. tuberculosis
using super-resolution microscopy combined with fluorescent
d-alanine analogues (FDAAs) (Botella et al. 2017b). They reported

that FDAAs are predominantly incorporated at one of the two
poles in M. smegmatis, whereas M. tuberculosis shows variation
in polar dominance depending on the stage in cell cycle. FDAAs
are also incorporated along the lateral wall upon damage due to
muramidase activity (Garcia-Heredia et al. 2018). Furthermore,
fluorescent intracellular membrane domain (IMD)-associated
protein reporters revealed that mycobacteria can spatiotempo-
rally coordinate its membrane domain in response to metabolic
requirements under different growth conditions (Hayashi et al.
2018). These recent reports on membrane compartmentalisation
of PG synthesis and its metabolism during the cell cycle in M.
tuberculosis will help to understand mechanisms allowing bac-
teria to escape host defence systems and to characterise drug
targeting steps that have remained elusive before.
Biosynthesis and maturation of PG
The nucleotide precursors of PG were first isolated from Staphy-
lococcus aureus in 1952 (Park 1952). Since then the various steps
involved in the biosynthesis of PG have been extensively studied
in a number of species. The PG biosynthetic pathway in M. tuber-
culosis, however, remains largely uncharacterised. In this work
we follow the generally accepted assumption that mycobacte-
ria synthesise PG precursors in a similar way to the model bac-
terium E. coli using enzymes that are homologous to the ones
found in the latter. However, as a slow-growing intra-cellular
pathogen with varied physiological states, mycobacteria have
PG enzymes that differ with respect to structure and regulation
compared to their counterparts in E. coli. This review attempts
to highlight these critical features. Details of the enzymes dis-
cussed throughout the review have been listed in Table S1 (Sup-
porting Information).
PG biosynthesis: the cytoplasmic steps
PG biosynthesis begins with the conversion of fructose-6-
phosphate to UDP-N-acetylglucosamine (UDP-GlcNAc), which
is catalysed by the enzymes GlmS, GlmM and GlmU (van
Heijenoort 2001a) (Fig. 2). Of these three enzymes, only
GlmU has been characterised in M. tuberculosis (Zhang et al.
2009; Jagtap et al. 2012). GlmU is a bifunctional acetyltrans-
ferase/uridyltransferase, which converts glucosamine-1-
phosphate (GlcN-1-P) to N-acetylglucosamine-1-phosphate
(GlcNAc-1-P) and subsequently catalyses the formation of
UDP-GlcNAc from GlcNAc-1-P and uridine triphosphate (Zhang
et al. 2009; Jagtap et al. 2012; Li et al. 2012). The reaction mech-
anism of the acetyl transfer by GlmU provides useful insights
for drug design, such as the conformational changes of the
protein structure upon interaction with the ligand (Craggs et al.
2018). Mycobacterium tuberculosis GlmU is trimeric in solution,
whereby each monomer folds into two distinct domains. The
N-terminal domain has a typical uridyltransferase fold based
on a dinucleotide-binding Rossmann fold and is similar to
that observed in the reported structure for GlmU from Strep-
tococcus pneumoniae (Sulzenbacher et al. 2001). The C-terminal
domain has a left-handed parallel β-helix fold forming the
acetyltransferase active site containing contributions from all
three subunits as is characteristic of other acetyltransferase
enzymes. However, the GlmU from M. tuberculosis is 6–8 fold
less active than that of GlmU from E. coli. Additionally, as M.
tuberculosis GlmU lacks free cysteines in the acetyl-CoA binding
site or any solvent-accessible cysteines elsewhere, it retains its
acetyltransferase activity even in the absence of reducing agents
and in the presence of a thiol-reactive reagent; both of which
Maitra et al. 551
render the E. coli enzyme inactive (Pompeo, van Heijenoort and
Mengin-Lecreulx 1998; Zhang et al. 2009). The crystal structure
of M. tuberculosis GlmU reveals a unique 30-residue extension
which forms a short helix at the C-terminus and is involved
in substrate binding (Jagtap et al. 2012). The C-terminus of the
enzyme is also involved in auto-regulation through phospho-
rylation by the serine/threonine protein kinase PknB (Parikh
et al. 2009). Dziadek et al. recently reported that GlmU interacts
with host immune factor IL-8, thereby increasing mycobacterial
attachment to neutrophils and facilitating infection (Dziadek
et al. 2016). How a cytosolic bacterial protein serves as a receptor
binding entity still requires clarification, however, the enzyme
has garnered interest for drug development. The enzyme is
essential for M. tuberculosis and it depletion results in severe
growth defects in vitro and reduced bacillary loads in mice
models (Soni et al. 2015). Further experiments showed that the
uridyl- and acetyl-transferase activities are individually essen-
tial for the pathogen thus inhibitors designed for either active
site would be effective. Also, as the synthesis of UDP-GlcNAc
occurs via a different enzymatic route in eukaryotes compared
to prokaryotes, any treatment designed against the latter would
be expected to have little effect on the former.
MurA (UDP-N-acetylglucosamine enolpyruvyl transferase)
and MurB (UDP-N-acetylenolpyruvyl-glucosamine reductase)
catalyse the conversion of UDP-N-acetylglucosamine (UDP-
GlcNAc) to UDP-N-acetylmuramate (UDP-MurNAc). MurA cataly-
ses the transfer of the enolpyruvyl moiety of phosphoenolpyru-
vate (PEP) to UDP-GlcNAc (De Smet et al. 1999; Xu et al. 2014).
MurB (Benson et al. 1993) is an NADPH dependent oxidoreduc-
tase that reduces the enolpyruvate moiety to d-lactate result-
ing in the formation of UDP-MurNAc. While mycobacteria and
Gram-negative bacteria like E. coli have a single copy of the murA
gene (Brown et al. 1995), a second transferase gene, murA2 occurs
as a duplicate copy in low G + C Gram-positive organisms such
as S. pneumoniae (Du et al. 2000). Interestingly, both copies per-
form the same function and can substitute for one another. The
purpose for such redundancy at the first committed step of PG
biosynthesis may be to accommodate differential regulation (Du
et al. 2000). UDP-MurNAc binds to E. coli MurA with high affin-
ity, suggesting an inhibitory or regulatory role in PG biosynthesis
(Mizyed et al. 2005). CwlM, an essential, cytoplasmic protein and
hypothetical PG amidase, is another regulator of MurA (Boutte
et al. 2016). Phophorylated CwlM increases the activity of MurA
nearly 30 fold. CwlM is phosphorylated only in the presence of
surplus nutrients, thereby revealing a possible mechanism by
which the pathogen correlates PG synthesis with the availabil-
ity of nutrients. Reducing the PG synthesis rate also has impli-
cations for drug susceptibility, therefore mechanisms which can
activate CwlM and MurA may help to target the ‘persister’ pop-
ulation of M. tuberculosis and reduce treatment duration, one of
the main causes of patient non-compliance. The regulatory role
of CwlM has since been investigated further and its influences
at later stages of PG synthesis have also been revealed and will
be discussed in the following section.
Homology modelling, molecular dynamics and molecular
docking studies on the mycobacterial MurA and MurB enzymes
revealed active site residues and helped identifying potential
inhibitors (Babajan et al. 2011; Kumar et al. 2011). The crystal
structure of M. tuberculosis MurB with FAD as the prosthetic
group was recently reported (Eniyan et al. 2018). MurB is a type
I UDP-GlcNAcEP reductase and shares conserved domains with
MurB from E. coli and Pseudomonas aeruginosa. The nicotinamide
and the enol pyruvyl moieties aligned well in M. tuberculosis
MurB upon superimposition with the other MurB structures and
 552 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5

Figure 2. An overview of PG biosynthesis, degradation and recycling pathways representing information currently known in Mycobacterium species. While there has
been plethora of information on the PG metabolism in other bacteria, identification of mycobacterial proteins involved in this process has been limited so far. ‘∗’ marks
enzymes that have not been experimentally established. This includes proteins that are putative or completely unknown and uncertainty on when the enzymatic
step occurs in the pathway. The redundancy in the PG hydrolases makes it difficult to assemble a comprehensive list within a figure and only the selected enzymes
discussed in this article are highlighted. Recycling of PG in mycobacteria is an especially underexplored area compared to other bacteria and may offer interesting
insight into homeostasis mechanisms within the cell.

show domain III to adopt a more open conformation on bind-
ing to the substrates. There are known inhibitors of MurB from
Gram-positive and Gram-negative bacteria, however, whether
these are active against M. tuberculosis MurB has not been inves-
tigated (Bronson et al. ; Yang et al. ).
The assembly of the peptide stem of the PG monomer unit
(UDP-MurNGlyc or UDP-MurNAc-pentapeptide) is catalysed by a
series of four essential and structurally related ATP-dependent
Mur ligases MurC, MurD, MurE and MurF which, in M. tubercu-
losis, catalyse the addition of l -Ala, d -Glu, mDAP and d -Ala-
d -Ala, respectively, with the concomitant hydrolysis of ATP to
ADP and inorganic phosphate (Fig. 2) (Basavannacharya et al.
2010a; Basavannacharya et al. 2010b; Munshi et al. 2013; Eniyan
et al. 2016). The mycobacterial Mur ligases catalyse amide bond
formation via a similar reaction mechanism to that seen in the
E. coli enzymes, i.e. the formation of an activated acylphosphate
derivative of UDP-MurNAc followed by a nucleophilic attack by
the amino group of the amino acid or dipeptide, ultimately
resulting in the formation of a peptide bond and release of inor-
ganic phosphate (Anderson et al. 1996; Falk et al. 1996; Liger et al.
1996; Tanner et al. 1996; Vaganay et al. 1996; Bertrand et al. 1999)
(Fig. 3E).
All ATP dependent Mur ligases (Mur synthetases) con-
tain common structural motifs despite their low amino
acid sequence identities. ATP-dependent Mur ligases can be
described as containing three distinct domains: (i) an N-
terminal Rossmann-type α/β fold domain that binds to the UDP-
substrate, (ii) a central ATPase domain (with a highly conserved
WalkerA-like motif, TG[T/S]XGK[T/S(2)]) and (iii) a C-terminal
domain that binds the appropriate amino acid (Smith 2006)
(Fig. 3).
The first amino acid of the peptide stem (l-Ala) is ligated
onto UDP-MurNAc by MurC. In M. leprae, l-Ala is replaced by
Gly. Interestingly, M. tuberculosis MurC is able to incorporate both
l-Ser and Gly into the growing peptide chain in vitro (Munshi
et al. 2013), although it is questionable whether the UDP sub-
strates containing these amino acids would be accepted by MurD
in the subsequent reaction. This substrate flexibility has also
been observed in some other organisms. Addition of d-Glu by
MurD is followed by incorporation of mDAP onto the growing
peptide chain by MurE. While over-expression of M. tuberculosis
MurE in E. coli and Pseudomonas putida does not affect growth,
a lytic effect was observed when the S. aureus MurE (which
incorporates l-Lys instead of mDAP) was overexpressed in E. coli
(Mengin-Lecreulx et al. 1999). In all mDAP-specific MurE ligases,
an arginine residue at position 416 binds to the free end of mDAP,
whereas this residue is replaced by an alanine or asparagine
in all l-Lys-specific MurE ligases (Dementin 2001). MurE from
M. tuberculosis contains three domains, similar to that reported
for E. coli. However, unlike its E. coli homologue, the mycobacte-
rial MurE coordinates a magnesium ion with the UDP-MurNAc-
l-Ala-d-Glu substrate (Basavannacharya et al. 2010a; Basavan-
nacharya et al. 2010b). The last two amino acids of the stem pep-
tide are added by MurF as the dipeptide d-Ala-d-Ala Interest-
ingly, MurF from E. coli accepts muropeptides with either mDAP
or l-Lys with similar efficiencies and its N-terminus is distinct
from the other Mur ligases (Yan et al. 2000).
The entire MurA-F pathway is unique to prokaryotes and
all corresponding genes are single-copy and essential in M.
tuberculosis (Sassetti, Boyd and Rubin 2003; Griffin et al. 2011;
DeJesus et al. 2017) making them attractive therapeutic tar-
gets. MurC-F share conserved residues and mode of action,
and are thus amenable to multi-target therapy. Development
of high-throughput activity assays as well as a one-pot assay
Maitra et al. 553
that screens for inhibitors of more than one enzyme in the Mur
pathway could aid drug discovery (Basavannacharya et al. 2010a;
Eniyan et al. 2016).
In addition to ATP-dependent Mur ligases, the enzymes
catalysing the formation of substrates for Mur ligase reactions
are also key components of PG biosynthetic machinery. MurI is
a glutamate racemase, which provides the d-Glu required for
PG synthesis. Unlike in Bacillus subtilis there is only one gene
coding for MurI in M. tuberculosis. The available crystal struc-
ture of MurI (Poen et al. 2016) and new inhibitors of the enzyme
(Prosser et al. 2016) strengthen its potential as druggable tar-
get in the pathogen. However, several issues question the value
of MurI as a drug target. Homogeneous MurI purifies as a sta-
ble dimer that is enzymatically inactive in vitro. It is presumed
that activator molecules such as UDP-MurNAc-l-Ala may disso-
ciate the dimer to generate the active enzyme as observed in
the case of E. coli MurI (Ho et al. 1995). This inability to extract
and purify an enzymatically active form of the enzyme from het-
erologous expression systems poses a significant bottleneck for
the development of high-throughput inhibitor screening assays.
Additionally, there is some ambiguity over its essentiality (Sas-
setti, Boyd and Rubin 2003; Harth et al. 2005; Griffin et al. 2011;
Morayya et al. 2015). Mortuza et al. identified a d-amino acid
transaminase in M. smegmatis that could complement the activ-
ity of MurI allowing murI mutants to grow in the absence of d-
Glu (Mortuza et al. 2018). Whether its homologue in M. tuber-
culosis also retains d-amino acid transaminase activity is not
known. The presence of enzymes with overlapping functions
indicates that resistance would develop easily against glutamate
racemase inhibitors. However, as TB chemotherapy is always a
combination, the rise of resistance may be controlled by admin-
istering a combination of drugs that work synergistically, such as
glutamate racemase inhibitors and d-cycloserine (David 2001).
mDAP is synthesised from aspartate in eight biosynthetic
steps (Usha et al. 2012). The enzymes in the pathway are essen-
tial and thus have the potential to become drug targets. Alpha-
ketopimelic acid inhibits the activity of DapA (dihydrodipicol-
inate synthase) in the micromolar range indicating that it is a
druggable target (Shrivastava et al. 2016). Mycobacterium tuber-
culosis DapE was found to be insensitive to the known DapE
inhibitor L-captopril (Usha et al. 2016). Finally, DapF in Chlamydia
trachomatis is bifunctional and involved in both mDAP synthesis
and the racemisation of glutamate (Liechti et al. 2018). Whether
the DapF from M. tuberculosis could also function as a d-Glu race-
mase is unknown. This essential enzyme converts ll-DAP into
mDAP. Although little variation was observed in the two-domain
α/β-fold of the apo structure of DapF from M. tuberculosis, when
compared with DapF from Haemophilus and Bacillus species, it
was noted that the M. tuberculosis DapF backbone is stabilised
via a tyrosine residue specific to mycobacterial DAP epimerases.
Any change of the interactions this residue is involved in would
destabilize DapF (Usha et al. 2009) thereby providing a rationale
for design of anti-tubercular specific DapF inhibitors.
The di-peptide substrate for MurF (d-Ala-d-Ala) is synthe-
sised by the alanine racemase Alr, followed by the d-Ala-d-Ala
ligase Ddl. Alr is a pyridoxal-5-phosphate-containing enzyme
(van Heijenoort 2001b) and its crystal structure shows a homod-
imer with the domains adopting a similar fold to that observed
in Alr structures from other species (LeMagueres et al. 2005).
However, the observed hinge angle between the domains is dif-
ferent for Alr from M. tuberculosis. The interest in Alr as a drug
target has been dampened by conflicting reports on its essen-
tiality in M. smegmatis and M. tuberculosis. While Milligan et al.
found deletion mutants require a steady supply of d-Ala for
 554 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5

Figure 3. (A), Cartoon representation of the crystal structure of MurE from M. tuberculosis. The UDP substrate binding domain 1 is shown in green, the central ATPase
domain 2 is shown in blue and the amino acid substrate binding domain 3 is shown in red. Image generated using PyMOL (PDB code: 2XJA). (B), Overlay of domain
2 from MurE from E. coli (PDB code: 1E8C, light blue) and M. tuberculosis (PDB code: 2XJA, dark blue). The P-loops are highlighted in cyan. (C), Interaction map of the
UDP-MurNAc substrate with MurE from M. tuberculosis. (D), Interaction map of ADP with MurE from M. tuberculosis. (E), The reaction mechanism of Mur ligases. The
‘AA’ in the UDP-MurNAc-AA substrate represents the peptide stems of varying lengths whereas the AAx .NH2 represents the incoming amino acid.

growth, insertion mutants that did not show growth defects in
the absence of d-Ala were selected for by Marshall et al. (Milli-
gan et al. 2007; Awasthy et al. 2012, Marshall et al. 2017). There-
fore, it is hypothesised that the function of Alr can be substi-
tuted by an unidentified transaminase that converts pyruvate
and d-Glu to d-Ala so Alr-specific inhibitors would likely have
no anti-mycobacterial effect.
Ddl catalyses the ATP-dependent ligation reaction to gener-
ate the dipeptide d-Ala-d-Ala. While only one copy of ddl was
identified in mycobacteria, two ddl genes are present in E. coli
(Zawadzke, Bugg and Walsh 1991). Ddl exists as a dimer, in which
each monomer contains three domains with the ligand binding
site formed at the intersection of these three domains (Bruning
et al. 2011). The apo form of Ddl from M. tuberculosis is unique
in that in the absence of substrate it adopts a closed confor-
mation that has to undergo a significant structural rearrange-
ment to bind to the substrates. The closed conformation of Ddl
offers many unique binding pockets adjacent to the active site
that may be useful for drug design. Ligand binding and catalytic
activity, therefore, requires significantly greater structural rear-
rangement compared to orthologues from other species such
as S. aureus. Biochemical analyses and classical enzyme kinetic
studies of Ddl from M. tuberculosis revealed the mechanism of
inhibition by d-cycloserine—knowledge that could be useful in
the design of future drugs (Prosser and de Carvalho 2013). Cur-
rently, d-cycloserine is known to target Ddl and Alr in M. tuber-
culosis, with strong inhibition observed in the case of Ddl and
weaker inhibition seen with Alr.
PG biosynthesis: the membrane-associated steps
A lipid carrier is essential for the transport of the hydrophilic PG
precursors across the hydrophobic cell membrane. Mycobacteria
differ from E. coli in terms of the molecule that serves as the lipid
carrier (decaprenyl phosphate versus undecaprenyl-phosphate)
and in the synthesis of this molecule. In fact, even within
mycobacteria there are marked differences within this path-
way. Based on sequence homology it is presumed that M. smeg-
matis genome codes for 4 to 10 prenyl diphosphate synthases,
whereas M. tuberculosis could contain up to 4 synthases (Crick
et al. 2000). Rv1086 and Rv2361 are two distinct prenyl diphos-
phate synthases, a ω,E,Z-farnesyl diphosphate synthase (Z-
FPPS) and a decaprenyl diphosphate (DecaPP) synthase, respec-
tively (Schulbach, Brennan and Crick 2000). These act sequen-
tially to synthesise decaprenyl diphosphate from isopentenyl
diphosphate and E-geranyl diphosphate (Schulbach, Brennan
and Crick 2000; Schulbach et al. 2001; Kaur, Brennan and Crick
2004). Rv1086 adds one isoprene unit to geranyl diphosphate,
generating the 15-carbon product (E,Z-farnesyl diphosphate).
Rv2361c then processively adds a further seven isoprene units
to E,Z-farnesyl diphosphate to generate the 50-carbon prenyl
diphosphate. The crystal structures of the enzymes show that
they are closely related, which partly explains how Rv2361 can
compensate for the deletion of Rv1086 (Schulbach et al. 2001;
Wang et al. 2008). The decaprenyl diphosphate, synthesised de
novo at the cytoplasmic side of the cell membrane or released by
the polymerisation of PG on the periplasmic face of the mem-
brane, must be dephosphorylated for the transfer of the acti-
vated sugar molecule from lipid I. The dephosphorylation of
decaprenyl diphosphate to decaprenyl phosphate is performed
by membrane-bound pyrophosphatases. In spite of multiple
undecaprenyl pyrophosphate phosphatases (Upp) identified in
E. coli (BacA, PgpB, YbjG and LpxT) the majority of this activity
(75%) is attributed to BacA (UppP); which also confers bacitracin
Maitra et al. 555
resistance in E. coli (El Ghachi et al. 2005; Bickford and Nick 2013).
PgpB and BacA interact with PBP1B (which polymerises PG) and
stimulate its activity by dephosphorylation of the released unde-
caprenyl pyrophosphate (Hernández-Rocamora et al. 2018). The
crystal structure of E. coli BacA shows a unique topology rais-
ing the possibility of its involvement as a lipid carrier flippase
in addition to being a phosphatase that could potentially accept
substrates from the cytoplasm as well as the periplasm (Work-
man, Worrall and Strynadka 2018). Despite the redundancy of
the phosphatases, deletion mutants lacking bacA show marked
phenotypic effects such as attenuation of virulence in Staphylo-
coccus aureus and Streptococcus pneumoniae in mice models, and
impaired biofilm formation in M. smegmatis (Chalker et al. 2000;
Röse, Kaufmann and Daugelat 2004). However, the M. tuberculo-
sis homologue of BacA was found to have no role in either acid
resistance or in virulence in mice models (Darby et al. 2011). This
suggests that there may be other enzymes in M. tuberculosis with
decaprenyl pyrophosphate phosphatase activity. The synthesis
and recycling of the lipid carrier is essential for the membrane-
bound steps of PG synthesis as discussed below and is a verified
target for antibiotics such as bacitracin (Siewert and Strominger
1967).
The first membrane step in the biosynthesis of PG involves
the phospho-N-acetylmuramoyl-pentapeptide transferase or
translocase MraY. MraY catalyses the transfer of the phospho-
MurNAc-pentapeptide moiety of UDP-MurNAc-pentapeptide
to the acceptor decaprenyl phosphate, yielding MurNAc-
(pentapeptide)-pyrophosphorylundecaprenol or lipid I. This pro-
ceeds via a covalent intermediate, which is formed upon nucle-
ophilic attack by an aspartate residue on the β-phosphate of
the UDP-MurNAc-pentapeptide resulting in the release of UMP
(Chung et al. 2013). While E. coli uses the undecaprenyl lipid,
M. tuberculosis solely uses the decaprenyl lipid, however, both
decaprenyl and heptaprenyl acceptors have been identified for
M. smegmatis (Takayama, Schnoes and Semmler 1973). Com-
prehensive biochemical or structural studies on mycobacterial
MraY have not been reported. However, active site mapping of
MraY from E. coli and comparison with MraY from other Gram-
negative and Gram-positive bacteria revealed five conserved
hydrophilic sequences located on the cytoplasmic side of the
membrane containing invariant amino acids (Al-Dabbagh et al.
2008). An X-ray crystal structure for MraY from Aquifex aeolicus
and a method to quantify MraY activity has provided further
insight into the active site residues (Chung et al. 2013). Improve-
ments of the biochemical assay to quantify MraY activity (Siri-
cilla et al. 2014) is a significant asset in the effort to develop
antimycobacterial agents targeting this protein.
Following the synthesis of lipid I, the N-acetylglucosamine
transferase MurG catalyses the transfer of GlcNAc from UDP-
GlcNAc to lipid I, yielding GlcNAc-MurNAc-(pentapeptide)-
pyrophosphoryl-decaprenol or lipid II.
MurT and GatD are responsible for the amidation of the d-
Glu residue in the peptide stem of both, lipid I and lipid II in
Staphylococcus aureus and Streptococcus pneumoniae (Figueiredo
et al. 2012; Munch et al. 2012). The arrangement of the murT and
gatD genes in an operon is conserved across various bacterial
families (Figueiredo et al. 2012) and has also been confirmed in
mycobacteria (Maitra et al., unpublished results). Crystal struc-
tures of the enzymes from S. aureus and S. pneumoniae reveal that
the pair form a stable, functional heterodimer wherein GatD is
responsible for the sequestration of the glutamine donor and its
subsequent deamination and MurT is a Mur ligase-like protein
that allows for the d-iGlu in the stem peptide to be amidated
to d-iGln and utilises an ATP in the process (Leisico et al. 2018;
 556 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
Morlot et al. 2018; Noldeke et al. 2018). While GatD is inactive in
the absence of MurT, the latter can functionally substitute the
role of MurE (Munch et al. 2012; Leisico et al. 2018). Amidation of
PG plays a crucial role in PG synthesis and maturation. It was
hypothesised that amidation could facilitate the translocation
of lipid II across the cytoplasmic membrane as a consequence
of the reduction of polarity (Munch et al. 2012). It has also been
reported that non-amidated lipid II is an inefficient substrate for
transpeptidation reactions in S. pneumoniae (Zapun et al. 2013).
Amidated muropeptides are recognized by PknB in M. tubercu-
losis and through the action of the kinase could regulate cell
growth, division and PG turnover (Mir et al. 2011).
The homologous murT/gatD genes in M. tuberculosis are
essential for in vitro growth of the pathogen (Sassetti, Boyd and
Rubin 2003). The essentiality of MurT/GatD and the requirement
for complex formation for activity may be exploited for design-
ing drug scaffolds.
Mycobacterial lipid II also contains amidated mDAP residues
(Mahapatra et al. 2005b). In Corynebacterium glutamicum and B.
subtilis, ltsA and asnB encode for the amidotransferases respon-
sible for the amidation of mDAP (Levefaudes et al. 2015; Dajkovic
et al. 2017). LtsA and AsnB belong to the growing family of glu-
tamine amidotransferases (such as GatD) whose members catal-
yse amide nitrogen transfer from glutamine to various spe-
cific acceptor substrates (mDAP in this case). Most of the inves-
tigations into the role of mDAP amidation have focussed on
the host immune response modulation of such modification.
Nucleotide-binding oligomerisation domain-containing protein
1 (NOD1) is a pattern recognition receptor that recognizes
MurNAc-tripeptide fragments containing mDAP and activates
the innate immune system (Girardin et al. 2003b). Amidation
of mDAP enables evasion of recognition by NOD1 by interact-
ing with residues away from the NOD1-leucine-rich repeat (LRR)
recognition region (Roychowdhury, Wolfert and Boons 2005;
Wolfert, Roychowdhury and Boons 2007; Vijayrajratnam et al.
2016; Wang et al. 2016). Unlike d-Glu amidation, the modifica-
tion of mDAP residue does not influence the pattern of insertion
of PG precursors into the cell wall or the overall cross-linking
levels (Levefaudes et al. 2015; Dajkovic et al. 2017). However, the
amidation of the mDAP residues in mycobacteria is essential for
the formation of LD-cross-links (Ngadjeua et al. 2018). Addition-
ally, the presence of zones of uncontrolled PG degradation in
the asnB mutant of B. subtilis suggests that mDAP amidation
controls the activity of hydrolytic enzymes such as autolysins
(Dajkovic et al. 2017). Once the modifications have been intro-
duced, lipid II is flipped to the outer face of the cytoplasmic
membrane.
The translocation of lipid II across the cytoplasmic mem-
brane has been subject to some controversy, with the role of
lipid II flippase having being contested between FtsW and MurJ.
The evidence supporting FtsW as the lipid II flippase comes
from an assay performed by Mohammadi et al. (Mohammadi
et al. 2011) on an in vitro reconstitution system, in which it was
observed that FtsW, and not MurJ, was able to translocate flu-
orescently labelled lipid II across the membrane of bacterial
membrane vesicles. It was later suggested by the same research
group that lipid II may be translocated via a pore-like structure
in FtsW (Mohammadi et al. 2014). Conversely, it has been sug-
gested based on genetic analyses and biochemical assays with
RodA, that proteins from the SEDS family may be glycosyltran-
ferases (GTases) rather than lipid II flippases (Meeske et al. 2016;
Emami et al. 2017). However, FtsW was not active as GTase in
another study in which it was also observed that FtsW, but not
MurJ, binds lipid (Leclercq et al. 2017). However, FtsW does in fact
have GTase activity, but only when in complex with its cognate
penicillin-binding protein (Taguchi et al. 2019).
The initial hypothesis for MurJ as the lipid II flippase was
based on its essentiality and the observation that PG synthe-
sis stops and precursors accumulate upon depletion of the murJ
gene in E. coli (Inoue et al. 2008; Ruiz 2008). A colicin M-based
cellular assay suggested that MurJ might flip lipid II across the
membrane (Sham et al. 2014). Native mass spectroscopy showed
that MurJ binds lipid II and that this binding is affected by car-
diolipin (Bolla et al. 2018). The crystal structures of MurJ from
Thermosipho afticanus (Kuk, Mashalidis and Lee 2017) and E. coli
(Zheng et al. 2018) confirmed similarity to MOP family of trans-
porters and further rationalised possible mechanisms by which
MurJ is likely to transport lipid II across the membrane. However,
until there is direct evidence with the mycobacterial protein, the
identity of the flippase in the pathogen remains uncertain.
Both PG precursor synthesis and their transport are regu-
lated by PknB-mediated phosphorylation of CwlM (introduced
earlier as regulating MurA activity). CwlM exists in two forms
within the mycobacterial cell. The non-phosphorylated CwlM
is membrane-associated, interacts with MurJ and presumably
enables it to transport lipid II (Turapov et al. 2018). The phos-
phorylated CwlM is cytoplasmic and interacts with FhaA and
MurA. The authors hypothesise that on binding to unpoly-
merised PG precursors in the periplasmic region, PknB under-
goes autophosphorylation and in turn phophorylates its sub-
strate CwlM, which dissociates from MurJ thereby reducing the
flipping of PG precursors. Direct phosphorylation of MurJ by
PknB was also proposed to produce the same outcome.
A number of compound classes bind to extracytoplasmic
lipid II and prevent PG polymerisation. Glycopeptides such as
vancomycin (MICMtb - 65 μg/mL) bind to d-Ala-d-Ala in lipid II
and nascent PG. Antimicrobial peptides such as nisin (MICMtb -
> 60 μg/mL), teixobactin (MICMtb - 0.125 μg/mL) and malacidin
(MICMtb - not determined) bind primarily to the pyrophosphate
moiety of lipid II and inhibit PG synthesis.
PG biosynthesis: linkage in mature PG
Once the precursors are translocated to the periplasmic space,
lipid II is used as the substrate for the subsequent polymerisa-
tion reactions and the formation of mature PG. The final stages
of PG maturation occurs in the periplasm, carried out by trans-
glycosylases and transpeptidases, which are often bifunctional
penicilloyl serine transferases or penicillin binding proteins
(PBPs) (Crick, Mahapatra and Brennan 2001; Sauvage et al. 2008),
resulting in the formation mature of PG chains and peptide with
DD-cross-links (Goffin and Ghuysen 1998). The class A and B
PBPs synthesize PG; whereas the class C PBPs are PG hydrolases
with either DD-endopeptidase or DD-carboxypeptidase activ-
ity. Both reactions provide the tetrapeptide substrates for LD-
transpeptidases. Class A PBPs (in mycobacteria PBP1 and PBP1A,
encoded by ponA1 and ponA2, respectively) consist of bifunc-
tional enzymes with an N-terminal glycosyltransferase and a C-
terminal DD-transpeptidase (penicillin-binding) domain (Maha-
patra et al. 2000; Bhakta and Basu 2002; Patru and Pavelka
2010). Class B PBPs (in mycobacteria PbpA and PbpB) are mono-
functional DD-transpeptidases (Dasgupta et al. 2006; Plocin-
ski et al. 2011). The glycosyltransferase domain catalyses the
transfer of the terminal N-glycolylmuramic acid residue from a
nascent glycan chain to N-acetylglucosamine of lipid II, releas-
ing decaprenol pyrophosphate (Scheffers and Pinho 2005; van
Heijenoort 2007). The C-terminal DD-transpeptidase domain
of PonA1 crosslinks the penultimate d-Ala to the mDAP of

another peptide (Filippova et al. 2016). The energy for the DD-
transpeptidation reaction stems from the cleavage of the d-Ala-
d-Ala peptide bond of the donor peptide resulting in the release
of the terminal d-Ala residue (Ghuysen 1991). PonA localises to
the poles and septum in mycobacteria and is essential in reg-
ulating cell length probably through its interaction with both
PknB and the peptidoglycan hydrolase RipA (Hett, Chao and
Rubin 2010; Kieser and Rubin 2014). The activities of PonA1
and PonA2 may be altered under different growth conditions as
mutants lacking one of the corresponding genes show differen-
tial response to cell wall antibiotics (Kieser et al. 2015).
The LD-cross-link was first identified by Wietzerbin in 1974
(Wietzerbin et al. 1974). Its physiological significance, however,
was revealed only in 2008 when Lavollay and colleagues dis-
covered that the PG of stationary phase M. tuberculosis almost
exclusively contained LD-cross-links (Lavollay et al. 2008). An
LD-transpeptidase, LdtMt1 , which catalysed the formation of LD-
cross-links, was identified and covalent adduct formation with
the β-lactamase drugs imipenem, meropenem and ertapenem
was demonstrated through mass spectroscopy (Lavollay et al.
2008). Subsequently, a second LD-transpeptidase, LdtMt2 was
identified by Gupta and colleagues, and loss of the ldtMt2 gene in
M. tuberculosis was shown to result in altered colony morphol-
ogy, loss of virulence and increased sensitivity to amoxicillin-
clavulanate (Gupta et al. 2010). In addition, LdtMt2 is critical for
growth in the absence of PonA1 or PonA2, suggesting PBPs and
LdtMt2 together support new cell wall synthesis during growth
(Kieser et al. 2015).
The influence of LD-cross-linking on mycobacterial phys-
iology and drug susceptibility was further investigated by
Baranowski and colleagues, making use of an M. smegmatis
strain lacking Ldts as a model (Baranowski et al. 2018). It was
observed that in the absence of LD-cross-links, the rigidity of
aging cell wall was compromised leading to the formation of
spherical blebs. The increased susceptibility of M. tuberculosis
to DD-transpeptidase targeting β-lactams was also confirmed
and it was demonstrated that inhibiting both DD- and LD-
transpeptidases with amoxicillin and meropenem respectively,
resulted in synergistic lowering of the MIC (Baranowski et al.
2018).
The two-step reaction of LD-transpeptidases begins with the
acylation of the enzyme by the third residue (l-chiral centre)
of the tetrapeptide donor, followed by deacylation of this acyl-
enzyme intermediate by the third residue (d-chiral centre) of
the adjacent acceptor stem resulting in the formation of a LD-
peptide linkage. The crystal structure of the extra-membrane
domain of LdtMt2 with a bound PG fragment (Erdemli et al. 2012)
showed that each monomer consists of two globular domains:
an N-terminal domain resembling an immunoglobulin domain,
and a C-terminal catalytic domain, consisting of a β sand-
wich with two mixed β sheets characteristic of the YkuD fold
(Bielnicki et al. 2006), with the two domains connected by a
short linker molecule. Structural, kinetic and calorimetric data
revealed a catalytic mechanism for LdtMt2 in which both acyl-
acceptor and acyl-donor substrates enter from the same side to
reach the catalytic site (Both et al. 2013; Triboulet et al. 2015; Bhat-
tacharjee et al. 2019).
The PG layer was long regarded as a network with mainly DD-
cross-links. However, the type of cross-links in PG can switch
to mainly LD-cross-links in response to changes in the environ-
ment and growth phase in Mycobacteria (Gupta et al. 2010) and
E. coli (Hugonnet et al. 2016). The shift to LD-cross-links allows
cells to grow in the presence of most β-lactam antibiotics that
target PBPs and not LD-transpeptidases, and the use of tetrapep-
tide donors by LD-transpeptidases allows for the generation of
Maitra et al. 557
peptide cross-links in the absence of de novo synthesis. These
features might facilitate the repair of damaged PG (Morè et al.
2019) or render PG resistant to hydrolytic enzymes (Lavollay et al.
2008), which could be vital during stationary phase or persis-
tence.
PG transpeptidases are verified but difficult targets for
antimicrobial drug discovery as (i) due to their redundancy,
many of these enzymes can substitute the function of another
and (ii) the drugs most commonly used to target them, the β-
lactams, are prone to inactivation by β-lactamases.
Drug development efforts targeting the LD-transpeptidases
have primarily focussed on further exploring the potential
of carbapenems, particularly meropenem, as Ldt inhibitors.
Shortly after the identification of LdtMt1 , Hugonnet and col-
leagues demonstrated that a combination of meropenem and
clavulanate was able to kill 13 different XDR strains of M.
tuberculosis with MIC values > 1 μg/mL (with one exception at
1.25 μg/mL) (Hugonnet et al. 2009). This built upon previous
work demonstrating that clavulanate can act as an irreversible
inhibitor of the M. tuberculosis β-lactamase BlaC (Hugonnet and
Blanchard 2007).
The direct targeting of LdtMt1 by carbapenems and
cephalosporins was investigated by Dubee and colleagues
and it was demonstrated that while both classes of β-lactamase
are able to form adducts with LdtMt1 , cephalosporins do so 7-
to 1000-times slower than carbapenems (Dubee et al. 2012).
Since then, Kim et al. showed that meropenem acts as a suicide
inhibitor of LdtMt2 by inducing conformational changes in the
protein; further informing drug design groups about inhibition
mechanisms for the protein (Kim et al. 2013).
Recently, in a small retrospective observational case study,
Paven and colleagues reported that of 18 patients with XDR-
TB who were treated with meropenem-clavulanate, 15 had
a ‘successful outcome’ (5 ‘cured’ and 10 ‘completed treat-
ment’), while two patients failed to complete the treatment
and one died (Payen et al. 2018). It is also notable that no
adverse effects were observed as a result of the treatment, sug-
gesting that with further study this may present a promis-
ing route to new β-lactamase-based TB treatment. However,
treatment of Mycobacterium bovis BCG with meropenem was
also reported to cause a significant increase in endogenous
ATP levels and suppression of this ATP spike using drugs
such as Bedaquiline attenuated the activity of the former
(Shetty and Dick 2018). Therefore, the inclusion of new drugs
into combination therapy poses complications of antagonistic
activity.
PG biosynthesis: orientation of mature PG
The orientation of PG has been a subject of debate and specu-
lation throughout the past decades. The classical depiction rep-
resents PG glycan strands as parallel to the bacterial membrane
surface as this provides better flexibility to accommodate cell
division (Koch 1998, Koch 2000; Vollmer and Holtje 2004; Schef-
fers and Pinho 2005; Vollmer 2008), however, others postulate
that PG is arranged perpendicular/orthogonal to the membrane
(Pink et al. 2000; Dmitriev et al. 2003; Dmitriev, Toukach and
Ehlers 2005; Meroueh et al. 2006). Using atomic force microscopy
the glycan strands of E. coli were seen running approximately
along the short axis of the cell, and the PG was more ordered
in rod shaped E. coli sacculi than spheroid shaped sacculi from
mutant strains (Turner et al. 2018). Our current understanding of
the order of PG glycan strands is largely derived from studies on
rod-shaped E. coli and B. subtilis, where new glycan strands are
inserted circumferentially forming new PG laterally along the
 558 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
side wall with the help of a protein complex called elongasome
(Dominguez-Escobar et al. 2011; Garner et al. 2011; Hussain et al.
2018). Whether a similar order of PG glycan strands is present in
mycobacteria is not known, as there are key differences between
mycobacterial PG biosynthesis and other rod-shaped bacterial
PG structure such as the incorporation of nascent PG at polar
ends, the presence of a high proportion of LD-cross-linkages and
the absence of MreB. The complexity of the mycobacterial cell
wall structure, lack of pure samples and the presence of abun-
dant secondary cell wall polymers make it challenging to deter-
mine its architecture.
PG biosynthesis: the dcw gene cluster
Although most of the genes involved in PG biosynthesis have
been identified in E. coli and their mycobacterial orthologues
are known, the differences in the structure of PG between
the organisms, however, warrants further investigation into
the products of these key genes. Since the sequencing the
genome of M. tuberculosis (Cole et al. 1998), a plethora of infor-
mation is available about the pathogen, aiding a more targeted
approach towards drug discovery and development. The major-
ity of PG biosynthetic genes in M. tuberculosis are located on
the division-cell wall (dcw) cluster along with genes required
for cell division (Munshi et al. 2013). This clustering of genes
is also observed in E. coli (Eveland, Pompliano and Anderson
1997), with the exception of the Rv2161c-Rv2160c-Rv2159c region
and the homologues present only in the slow growing mycobac-
terial species such as M. tuberculosis (Fig. 4). The clustering
of cell division genes (ftsZ, ftsQ and ftsW) and PG biosynthe-
sis genes in the dcw operon are indicative of a complex and
spatio-temporal coordination between cell division proteins and
cell wall precursor synthesis enzymes (Mingorance, Tamames
and Vicente 2004; Vicente et al. 2006; Munshi et al. 2013; Carel
et al. 2014).
Degradation and remodelling of PG
PG is a dynamic structure that is constantly synthesised and
remodelled. Cell growth, division and septation require the
co-ordination of the enzymes required for both PG synthe-
sis and hydrolysis. PG hydrolases (PGH) form a vast group
of enzymes including glycosidases, amidases, endopeptidases
and carboxypeptidases, which together are capable of fully
digesting the PG polymer into soluble fragments. They can act
by cleaving bonds within polymeric or soluble fragments of
PG (Loessner et al. 2002; Vollmer et al. 2008), and hence may
have a periplasmic, membrane-associated or cytoplasmic loca-
tion (Holtje 1998). Often called autolysins, they are capable of
lysing whole cells when deregulated. Due to their selective
ability to cleave PG, the hydrolases have been implicated in a
plethora of cellular processes including cell growth, PG matu-
ration, PG turnover, cell division, sporulation, resuscitation and
pathogenicity (Goodell and Schwarz 1985; Heidrich et al. 2002;
Morlot et al. 2010; Lee and Huang 2013). Due to their central role
in the processes outlined above these enzymes offer a unique
opportunity to alter the permeability of the mycobacterial cell
wall and so make M. tuberculosis and related pathogens more
susceptible to drug treatment. In support of their importance
in regulation of bacterial physiology Machowski et al. recently
identified multiple homologues of these PG remodelling genes
in M. tuberculosis (Machowski et al. 2014).
Glycosidases such as N-acetyl-β-d-glucosaminidases and N-
acetyl-β-d-muramidases cleave the polysaccharide backbone
of PG, while N-acetylmuramyl-l-Ala amidases hydrolyse the
amide bond between MurNAc and l-Ala of the stem peptide.
Lytic transglycosylases are, by virtue of their enzymatic reac-
tion, not ‘hydrolases’ but they are usually discussed amongst
the PG hydrolases. Unlike other N-acetyl-β-d-muramidases like
lysozyme the lytic transglycosylases catalyse the formation of
a 1,6-anhydromuramyl ring by an intramolecular transglycosy-
lation reaction in the MurNAc residue of the released PG frag-
ment. Different forms of endopeptidases and carboxypeptidases
are capable of hydrolyzing amide bonds either in the stem pep-
tide or cross-linkages of PG (Scheurwater, Reid and Clarke 2008;
Vollmer et al. 2008).
The presence of three major autolysins, N-MurGly-l-Ala ami-
dase, an amino peptidase and an endopeptidase in M. smegmatis,
was first reported in 1977 (Kilburn and Best 1977) followed later
by partial characterisation of a β-glycosidase from M. phlei (Li
et al. 1999).
DacB2 from M. tuberculosis is a DD-carboxypeptidase (Class
C PBP) that removes d-Ala residues at position 5 of pentapep-
tides (Kumar et al. 2012). The overexpression of dacB2 results in
altered morphology and defective biofilm formation, while its
deletion leads to increased intracellular survival of the mutant
in THP-1 cells (Bourai, Jacobs and Narayanan 2012). DacB2 was
also reported to possess DD-endopeptidase activity (Baranowski
et al. 2018). Thus, it can affect cell morphology by hydrolysing
the DD-cross-links as well providing tetrapeptide substrates for
LD-transpeptidases by removing the terminal d-Ala residue.
Early investigations by Votyakova (Votyakova et al. 1994) and
Mukamolova (Mukamolova et al. 1998b; Mukamolova et al. 1999;
Mukamolova et al. 2002) in Micrococcus luteus suggested the pres-
ence of a resuscitation promoting factor (Rpf) with lytic transg-
lycosylase activity responsible for the release of PG fragments
(Fig. 2). This ground-breaking discovery led to the identifica-
tion of five Rpfs (rpfA-rpfE) in Mycobacterium spp. (Cole et al.
1998; Mukamolova et al. 1998a; Sutcliffe and Harrington 2004),
which significantly enhanced our understanding not only of
the remodelling of cell wall PG during daughter cell formation,
but also the processes behind dormancy and reactivation of
bacilli during active and latent TB. Although all five Rpfs are
dispensable for mycobacterial growth, the lack of three of the
five rpfs leads to avirulence in mice and an inability to resus-
citate from dormant bacilli (Downing et al. 2005). Furthermore,
the lack of spontaneous reactivation of the rpf deletion mutants
reveals a synergy between them, suggesting that their func-
tion cannot be substituted by another enzyme (Downing et al.
2005, Kana et al. 2008). Interestingly, an Rpf-related motif iden-
tified in mycobacteriophages is involved in PG degradation to
facilitate phage infection during the stationary phase (Piuri and
Hatfull 2006). This suggested that the Rpf-related motif was
important for the breakdown of PG with extensive ld-cross-links
such as those encountered in dormant M. tuberculosis. The Rpfs
have detectable sequence homology with lysozymes and lytic
transglycosylases (Cohen-Gonsaud et al. 2004; Mukamolova et al.
2006). The C-terminal domain of RfpB from M. tuberculosis com-
prises a PG binding cleft with two parts. One forms a compact
lysozyme-like fold resembling that of c-lysozyme, whereas the
other resembles soluble lytic transglycosylase proteins such as
E. coli Slt70 (van Asselt, Thunnissen and Dijkstra 1999; Cohen-
Gonsaud et al. 2005). Although the catalytic domains in RpfE and
RpfC are similar to those in RfpB (Chauviac et al. 2014; Mavrici,
Prigozhin and Alber 2014), the substrate-binding site in RpfE is
more positively charged than that in RpfB. This suggests that
these Rpfs either function optimally at different pH values or
cleave different micro-domains of PG.

Figure 4. Conservation in genomic organization of dcw gene clusters. Here, we compare the differences in the gene content of the dcw clusters of fast and slow growing
mycobacteria with that of E. coli.
Due to the network-like structure of PG, its enlargement
during cell growth and division requires the synergistic activ-
ity of both biosynthetic and hydrolytic enzymes for the incor-
poration of new glycan strands into the growing PG sacculus.
PG hydrolases can destroy the integrity of the sacculus; hence
their activity must be controlled in the cell to avoid autolysis.
There is mounting evidence that the biosynthetic and hydrolytic
enzymes interact with each other to form complexes capable
of synergistic breaking and making of bonds in PG and regu-
lating activity. In E. coli, Slt70 (a soluble lytic transglycosylase)
forms a complex with PBP 3 (a PG transpeptidase) and PBP 7/8
(a DD-endopeptidase) and both enzymes synergistically degrade
PG (Romeis and Holtje 1994). Slt70 also interacts with PBP1B,
PBP1C and PBP 2 (von Rechenberg, Ursinus and Holtje 1996). The
lytic transglycosylase MltA interacts with PBP1B (a bifunctional
synthase) and MipA (MltA-interacting structural protein). MtlB
interacts with PBP1B, PBP1C and PBP 3 (von Rechenberg, Ursi-
nus and Holtje 1996; Vollmer, von Rechenberg and Holtje 1999).
In Mycobacterium spp. the resuscitation promoting factors RpfB
and RpfE interact with the endopeptidase RipA at the septum of
dividing bacteria (Hett et al. 2007; Hett and Rubin 2008). An inter-
action between SltB1 and PBP2 has been observed in P. aeruginosa
(Legaree and Clarke 2008; Nikolaidis et al. 2012).
Mycobacterial RipA is a d-Glu-mDAP endopeptidase that
cleaves between positions 2 and 3 of the stem peptide (Both,
Schneider and Schnell 2011). RipA is essential in M. tuberculosis
(Sassetti, Boyd and Rubin 2003) and its depletion in M. smegmatis
causes severe growth inhibition (Hett and Rubin 2008). Unregu-
lated, RipA behaves like an autolysin in M. tuberculosis and M.
smegmatis, resulting in spherical cells that eventually lyse (Chao
et al. 2013). RipA, belonging to NlpC/P60 family, has a Cys-His-
Glu catalytic triad residing in the C-terminal catalytic domain,
(Ruggiero et al. 2010) and is regulated by proteolytic cleavage by
MarP (Botella et al. 2017a). RipA becomes a lethal autolysin in
the absence of protein interactions that are necessary to con-
trol its proteolytic activation (Chao et al. 2013). RipA-mediated
cleavage in PG and its activation is more robust in the fast-
growing environmental species M. smegmatis compared to the
slow growing M. tuberculosis or M. bovis BCG—presumably a con-
sequence of the varying demands for PG hydrolysis required
Maitra et al. 559
at differing growth rates. RipB has unique features in its N-
terminal segment, which runs across the catalytic core domain
and forms a helix instead of the β-hairpin loop seen in the
homologous module of RipA (Both, Schneider and Schnell 2011).
These enzymes are not essential and so are not attractive drug
targets.
N-acetylmuramyl-l-Ala amidases hydrolyse the bond
between MurNAc and the peptide portion of the PG and have
been implicated in PG degradation and turnover, cell separation,
antibiotic resistance and spore formation (Heidrich et al. 2001;
Vollmer et al. 2008). PG amidases fall into two groups; one
containing the amidase 2 domain and other the amidase 3
zinc-dependent domain (which includes the E. coli AmiA, AmiB
and AmiC and B. subtilis CwlB and CwlC enzymes). The first
autolysin identified in M. tuberculosis was an amidase known
as CwlM (Rv3915), which has since been established to have a
regulatory role in PG synthesis and maintenance as mentioned
in preceding sections. Another homologue (Rv3717) of the
E. coli PG amidases recently identified in M. tuberculosis was
found to lack a PG-binding domain and is active on muramyl
dipeptide, but not on polymerised PG (Kumar et al. 2013;
Prigozhin et al. 2013). The catalytic function identified by the
structural analysis of Rv3717 indicates that the catalytic pocket
of this enzyme is shaped as a blind tunnel and accommodates
only PG monomers, as peptide cross-links in polymerised PG
cannot enter the tunnel. It uses an overall positive charge on its
surface to bind to its substrate and regulates its activity through
an auto-regulatory β-hairpin (Kumar et al. 2013). As this is a
cytosolic protein the activity of this enzyme depends on the
action of another PG hydrolase cleaving the cross-links of PG
fragments released from the cell wall and transported back into
the cytoplasm. This indicates a possible involvement for Rv3717
in PG recycling (Prigozhin et al. 2013).
Transport and recycling of degraded PG components
A study of turnover rates in E. coli led to the serendipitous dis-
covery of PG recycling (Goodell and Schwarz 1985). The extent of
PG fragments released from the cell wall of growing bacteria, is
in the range of 25%–50% in Gram-positive organisms such as B.
 560 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
subtilis (Mauck, Chan and Glaser 1971) and E. coli (Chaloupka and
Strnadova 1972; Goodell 1985; Goodell and Schwarz 1985). How-
ever, E. coli recycles about 90% of the turnover products, releas-
ing only 0%–5% of the PG material into the culture supernatant.
Until recently, it was believed that PG recycling is a characteris-
tic of Gram-negative bacteria only (Herve et al. 2007). However,
the identification of orthologues of recycling enzymes in B. sub-
tilis and subsequent muropeptide recovery studies have uncov-
ered evidence of PG recycling in Gram-positive organisms as well
(Reith and Mayer 2011; Kluj et al. 2018). Apart from the obvi-
ous advantages associated with conservation of resources and
energy, PG recycling also plays a significant role in monitoring
the integrity of the cell wall, cell signalling and antibiotic resis-
tance via β-lactamase induction (Jacobs et al. 1994). Additionally
PG recycling plays a crucial role in the suppression of the innate
immune response, by effectively recovering cell wall muropep-
tides that would otherwise stimulate immunity, via proteins that
recognise PG (Boudreau, Fisher and Mobashery 2012; Johnson,
Fisher and Mobashery 2013).
Although there is little knowledge about PG recycling in
M. tuberculosis, some similarities have been observed with
the well-studied system in E. coli. Recycling of PG precursors
entails lateral wall and septal PG degradation to form GlcNAc-
(1→4)-1,6-anhydro-MurNAc-peptide monomers, GlcNAc-(1→4)-
1,6-anhydro-MurNAc dissacharides and free peptides (Goodell
1985). Pulse-chase experiments with tritiated mDAP provided
evidence to suggest that many of the cell wall precursors were
recycled (Goodell and Schwarz 1985). AmpG from E. coli, also
required to induce AmpC (a β-lactamase), was identified as
the permease transporting the primary products of the action
of lytic transglycosylases on PG into the cytoplasm (Korfmann
and Sanders 1989; Jacobs et al. 1994; Cheng and Park 2002).
Although Gram-positive organisms lack an AmpG homologue,
the presence of muropeptides with various length peptide side
chains (tetrapeptides, dipeptides) (Mahapatra et al. 2008), as well
as inducible β-lactamases (Flores, Parsons and Pavelka 2005),
suggests that an equivalent pathway for the recycling path-
way for PG exists in mycobacteria. Once the muropeptides have
been transported back to the cytoplasm, various amidases and
epimerases, which are non-essential for growth, further process
them to products that can be utilised for de novo PG precursor
synthesis (Park and Uehara 2008).
Homologues of the enzymes involved in the import, break-
down and recycling of muropeptide in the cytoplasm are yet to
be identified in M. tuberculosis. In the preceding section we dis-
cussed the various enzymes involved in degradation and remod-
elling of the cell wall PG. Recent work by Moynihan et al. pro-
vided evidence for the uptake of muropeptide fragments in M.
tuberculosis and M. bovis BCG; where they show cleavage of stem
peptide followed by disaccharide cleavage by LpqI (NagZ) and
finally lactyl-ether removal from the MurNAc moiety (Moyni-
han et al. 2019). NagA is involved in PG recycling through the
deacetylation of GlcNAc-6P in a number of other organisms,
including E. coli, B. subtilis, S. aureus and Streptomyces coelicolor.
A putative nagA gene in an operon distinct from other bacterial
organisms, has been identified in M. tuberculosis (Ahangar et al.
2018). Mycobacterium tuberculosis nagA is essential and encoded
along with genes involved in PG biosynthesis and carbohydrate
uptake, indicating that it may play a role in the recycling of
GlcNAc6P derived from the breakdown of PG.
Structural characterisation of NagA reveal it to be a dimer,
with each unit organised in two domains with active site pockets
occurring in domain I (Ahangar et al. 2018). The crystal structure
reveals the binding of the GlcNAc-6P substrate to M. smegmatis
NagA and indicates that the active site does not undergo any
major structural change on binding the substrate. However, two
loop regions demonstrate a capacity for conformational flexibil-
ity and an importance in facilitation of substrate binding, pro-
viding opportunity for designing inhibitor molecules (Ahangar
et al. 2018). Biochemical characterization shows a clear prefer-
ence of NagA for GlcNAc6P over other amino sugar analogues.
The stringent recognition of this fragment is hypothesised to
ensure the integrity of PG recycling (Ahangar et al. 2018).
The essentiality of nagA makes it a feasible drug target.
Eukaryotic homologues of the enzyme have low sequence iden-
tity and accommodate larger, glycolyl substituted substrates.
This property makes the two homologues distinct, however,
small inhibitors acting on M. tuberculosis NagA may be accepted
by mammalian homologues thereby giving rise to cytotoxicity
(Bergfeld et al. 2012).
Recycling of stem peptides is facilitated by murein peptide
ligase (Mpl) in Gram-negative organisms. Mpl ligates di-, tri- and
tetrapeptide stems onto UDP-MurNAc in an ATP-dependent step
thereby replacing the functions of MurC-F playing an impor-
tant role in the transition of cell towards stationary phase, as
observed in E. coli (Talukder et al. 1996). An mpl homologue has
not been identified in Gram positive bacteria or in any mycobac-
terial species.
Role of PG in cell signalling
PG fragments released from the cell envelope during growth
and cell division play important roles in cellular signalling and
inter-bacterial communication (Humann and Lenz 2009) (Fig. 5).
Rpfs produce PG fragments that lead to the resuscitation of dor-
mant mycobacteria (Mukamolova et al. 1998a; Mukamolova et al.
1999; Nikitushkin et al. 2013). RpfC and RpfD are restricted to
pathogenic and environmental mycobacteria and are secreted
into the culture medium, unlike RpfA, RpfB and RpfE, which
are membrane-bound (Machowski et al. 2014). This implies that
specific PG products are likely to act as messengers, which can
act as a signal to reawaken dormant cells (Mukamolova et al.
1998a; Zhang et al. 2001; Machowski et al. 2014). The release of
mDAP-containing muropeptides during growth triggers sporula-
tion in B. subtilis and by binding of the serine threonine protein
kinase (STPK) PrkC to the PG fragment (Shah et al. 2008). It is
possible that in mycobacteria the PG fragments released by Rpfs
bind PknB, triggering a cascade of signals leading to resuscita-
tion of the bacteria (Fig. 5). Furthermore, Russell-Golman et al.
also showed that mycobacterial double mutants of RpfA and
RpfE show a reactivation-deficient and attenuated phenotype
(Russell-Goldman et al. 2008), suggesting that Rpfs serve as vir-
ulence factors (Romano et al. 2012). In addition, Rpfs induce ele-
vated levels of cytokines in cultures from individuals with latent
TB (Riano et al. 2012).
Although PG has been the subject of intense study over the
last number of decades, the changes that occur when a cell
enters and exits stationary phase are not well understood. In
some bacteria, d-amino acids are incorporated into the PG poly-
mer in the stationary phase, which affects the amount of PG per
cell (Lam et al. 2009; Cava et al. 2011). d-Ala inhibits the germi-
nation of spores in B. subtilis (Chesnokova et al. 2009). d-Met and
d-Leu were the most prominent non-canonical d-amino acids
incorporated into the PG during the stationary phase in Vibrio
cholerae (Lam et al. 2009).
The ability of a host to recognise bacterial components
and trigger an innate immune response is the first line of
defence against infection. The fact that eukaryotes do not
 Maitra et al. 561

Figure 5. The role of PG and its breakdown products in signalling. The digestion of PG by hydrolases generates products like muramyl dipeptide and muramyl tripeptide
that are key signalling molecules to resuscitate bacterial cells and may induce β-lactamase. PG fragments are encountered by the host immune surveillance systems
either extracellularly (e.g. in serum) or are phagocytosed triggering various pro-inflammatory host immune responses.
 562 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
produce PG has led to the evolution of several strategies to
detect PG or PG fragments (Humann and Lenz 2009; Otten
et al. 2018). The germline-encoded pattern-recognition recep-
tors (PRRs) are expressed mainly on immune cells and recognise
pathogen-associated molecular patterns (PAMPs) and mediate
the production of immunoregulatory cytokines such as tumour
necrosis factor (TNF) and type I interferons (IFNs) (Medzhi-
tov and Janeway 1997). Toll-like receptors (TLR) are activated
by immunomodulatory lipoproteins in mycobacteria such as
lipomannan (LM), lipoarabinomannan (LAM), mannosyl capped
LAM (ManLAM) or phosphoinositide capped LAM (PILAM) and PG
components present on the cell surface or endosome/lysosome
membranes (Gilleron, Quesniaux and Puzo 2003; Quesniaux et al.
2004; Singh et al. 2012; Esin et al. 2013; Xie et al. 2013). The cyto-
plasm of the cell is also under scrutiny for PAMPs that breach the
cell membrane. These PAMPs are recognised by constitutively
expressed Nod-like receptors (NLR) like NOD, NALP and NAIP.
They consist of the NACHT or NOD oligomerisation domain and
a motif similar to the microbial sensing leucine-rich repeat (LRR)
receptors present in the TLRs. While NOD1 specifically recog-
nises the γ -d-glutamyl-mDAP peptide of PG fragments, NOD2
and NALP3 are activated by intracellular muramyl dipeptides
(Mukamolova et al. 1998a; Zhang et al. 2001; Chamaillard et al.
2003a; Girardin et al. 2003a; Chamaillard et al. 2003b; Uehara et al.
2006; Humann and Lenz 2009; Pandey et al. 2009; Killick et al.
2013; Machowski et al. 2014). Thus, any changes in the structure
of PG could impair the innate immune responses of the host.
Targeting PG biosynthesis to tackle antibiotic resistance
(R)
The ‘Golden Age’ of antibacterial drug discovery appears to be
over: since the 1990s there has been a void in the development
of new drugs for the treatment of bacterial infections (Silver
2011). More specifically, for TB the current frontline treatment
regimen is a cocktail of four drugs (isoniazid, rifampicin, pyraz-
inamide and ethambutol) that was first introduced in the 1960’s
(Zumla, Nahid and Cole 2013). The emergence of MDR- and XDR-
TB strains, however, have reinvigorated the search for antimy-
cobacterial agents with novel modes of action (Young et al. 2008).
Recent progress in this area includes the approval of Bedaquiline
(Sirturo) (Mahajan 2013; Koul et al. 2014) and Delamanid (Del-
tyba) (Ryan and Lo 2014).
Pathways involved in the biosynthesis of cell wall PG have
historically been recognised as important drug targets. This is
demonstrated by the fact that a significant number of the antibi-
otics in current clinical use, primarily β-lactams (e.g. penicillins
and cephalosporins) and glycopeptides, act by inhibiting the
later stages of PG biosynthesis (Fig. 2). Mycobaterium tuberculosis
is considered resistant to many β-lactams, due to a combina-
tion of the following factors: the thick, complex and hydropho-
bic nature of its cell envelope, the presence of one or more
active β-lactamases (Nampoothiri et al. 2008) and low-affinity
PBPs (Chambers et al. 1995). A number of clinically useful TB
drugs, however, do target cell wall processes. These include: (i)
the first line anti-TB drug isoniazid which inhibits MA biosyn-
thesis (Zhang et al. 1992), (ii) ethambutol, another frontline drug,
which inhibits AG synthesis (Mikusova et al. 1995) and (iii) a sec-
ond line drug d-cycloserine which is a broad spectrum antibi-
otic that inhibits PG biosynthesis via inhibition of Alr and Ddl
(Zhang 2005). Delamanid, like isoniazid, also targets genes in MA
biosynthesis (Ryan and Lo 2014); and Pretomanid (PA-824), an
antitubercular drug under regulatory review by the FDA, has also
been shown to affect not only respiratory genes, but also tran-
scription of genes responsive to known cell wall inhibitors like
isoniazid (Manjunatha, Boshoff and Barry 2009). The efficacy of
these drugs highlights the essentiality of the mycobacterial cell
wall PG and highlights the opportunity to develop novel chem-
ical entities, which target enzymes involved in the synthesis of
the mycobacterial cell envelope (Feng and Barletta 2003; Silver
2006; Prosser and de Carvalho 2013). Further research to iden-
tify the unique features of the mycobacterial PG enzymes with
respect to their structure, function and regulation can advance
the development of narrow-spectrum, targeted antibiotics.
To date the early stages of PG biosynthesis have been poorly
exploited as antibacterial targets, and so present a significant
opportunity for the development of the next generation of
antimycobacterial drugs. In particular, the enzymes involved in
the cytoplasmic stage of the PG biosynthetic pathway are cur-
rently considered as new targets for drug discovery. This inter-
est, however, has yet to translate to the development of a clin-
ically useful molecule. Although inhibitors against MurA and
MurB have been reported previously (Baum et al. 2001; Bronson
et al. 2003; Yang et al. 2006) (1, 2, 3, Fig. 6.) herein, we will review
the progress made to date towards the development of inhibitors
targeting the ATP-dependent Mur ligases MurC–F involved in the
cytoplasmic phase of PG biosynthesis.
The Mur ligases are found in all clinically significant bacte-
rial pathogens, and so inhibitors would be expected to be broad
spectrum. In addition, (i) there are conserved structural and
sequence motifs and 3 binding sites in MurC-F against which
antimicrobial agents can be designed (e.g. MurE Fig. 3) and (ii)
partial inhibition at multiple steps of PG synthesis (various Mur
enzymes) compared to inhibition of a single enzyme in the path-
way may provide a more efficient route to obtaining bacterici-
dal activity and decrease the likelihood of development of resis-
tance. As a result, pan or multi-target inhibitors are conceivable
and these may provide a better approach to obtaining robust
antibacterial activity.
Adding to the appeal of the ATP-dependent Mur ligases as
targets for drug design, many X-ray crystal structures have been
solved for proteins from different bacterial species (Bertrand
et al. 1999; Deva et al. 2006; Basavannacharya et al. 2010a). How-
ever, only the X-ray crystal structure of MurE from M. tubercu-
losis is currently available (Basavannacharya et al. 2010a; Basa-
vannacharya et al. 2010b). Homology models of MurC and MurD
from M. tuberculosis and MurC-F from M. leprae have been built
based on sequence identification with the corresponding homo-
logue from either E. coli or H. influenzae (Anuradha et al. 2010;
Arvind et al. 2012; Shanmugam, Anbazhagan and Natarajan
2012); Shanmugam and Natarajan 2012). Docking of native sub-
strates to these models aligned well with known X-ray crystal
structures and good homology was observed between conserved
binding site residues (Anuradha et al. 2010; Arvind et al. 2012;
Shanmugam and Natarajan 2012). In silico docking of naphtha-
lene sulphonamide inhibitors recapitulated the binding mode
observed in X-ray crystal structures and highlighted the impor-
tance of the Glu moiety as a driver of affinity for MurD (Arvind
et al. 2012). One may expect that inhibitors of Mur ligases
from other species may also inhibit the mycobacterial enzymes.
These models, therefore, provide a good starting point for the
design of novel ligands, however, experimental validation of the
predicted structures is currently lacking.
The progress made to date in the design of Mur ligase
inhibitors has recently been reviewed, although it is notewor-
thy that activity data on M. tuberculosis are woefully missing (Sil-
ver 2011; Perdih et al. 2014; Sangshetti et al. 2017). The inhibitors
discovered to date fall primarily into three classes: (i) transi-
tion state analogues, which contain for example phosphinate

Figure 6. Structures of representative inhibitors of MurA-F with inhibitory data.
or sulphonamide functional groups which mimic the tetrahe-
dral transition state formed during the enzyme-catalysed reac-
tion (Fig. 3) (Marmor et al. 2001; Reck et al. 2001), (ii) phos-
phate/diphosphate mimics which typically contain a hetero-
cyclic moiety (e.g. rhodanine, thiazolidindione) (Barreteau et al.
Maitra et al. 563
2012) and (iii) inhibitors that bind to the ATP binding site
(Hameed P et al. 2014). A number of core functional groups are
repeatedly found in inhibitors of MurA to F, particularly those
which act as mimics of the tetrahedral transition state or are
phosphate bioisosteres. The majority of inhibitors developed
 564 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
to date have arisen from HTS ( in vitro or in silico) followed by
structure-guided design, however, natural products have also
been the source of a number of inhibitors (Guzman et al. 2010;
Guzman et al. 2013b).
The majority of published MurC inhibitors are either phos-
phinate transition state analogues (4, Fig. 6) (Marmor et al. 2001;
Reck et al. 2001) or l-Ala analogues (El Zoeiby et al. 2003). To the
best of our knowledge, however, antibacterial activity has not
been demonstrated for either class, which is perhaps unsurpris-
ing as the above compounds are very polar and are unlikely to
be membrane permeable. The most promising MurC inhibitors
identified to date were reported in a study by Astra Zeneca in
which a series of potent pyrazolopyrimidines were developed
(Fig. 6, 5). Compound 5 was proposed to bind into the ATP bind-
ing site and encouragingly, also demonstrated bactericidal activ-
ity, albeit against efflux deficient mutants (MIC = 3.1 and 1.5
μM against E. coli and P. aeruginosa respectively). Importantly, 5
also demonstrated selectivity by not inhibiting a panel of human
kinases. From mutation resistance maps 5 was suggested to be
selective for MurC over the other Mur ligases, however, biochem-
ical data to test for inhibition of MurD-F is lacking. Most signifi-
cantly, the mode of action of the pyrazolopyrimidines was con-
clusively demonstrated to be as a result of inhibition of MurC.
This is a particularly noteworthy advance, as it is the first exam-
ple in which an antibacterial activity of a Mur ligase inhibitor has
been demonstrated to be due to target engagement (Hameed P
et al. 2014).
MurC phosphinate transition state analogues are also potent,
sub-nanomolar inhibitors of MurD. Replacement of the uridine
core with an aryl sulphonamide as a phosphate mimic afforded
the simplified inhibitor 6 (Fig. 6), which retained an IC50 in the
low micromolar range against MurD and was more amenable to
optimisation by medicinal chemistry (Strancar et al. 2006). The d-
glutamate moiety or conformationally restrained analogues are
present in the majority of published inhibitors, as they are key
to potent inhibition of MurD (7, 8, Fig. 6.) (Barreteau et al. 2012).
Another class of MurD inhibitors feature a benzylidene rhoda-
nine moiety, which occupies the uridine diphosphate binding
site (7, Fig. 6). Although 7 is a low micromolar inhibitor of the E.
coli enzyme, it inhibits the mycobacterial enzyme with an IC50 of
200 μM, highlighting the challenges associated with the design
of broad spectrum agents (Barreteau et al. 2012).
A number of the reported MurD inhibitors also inhibit MurE
and, again, contain phosphinate groups (or sulphonamides)
that mimic the tetrahedral intermediate formed during the
enzyme-catalysed reaction (Tanner et al. 1996; Strancar et al.
2007). Alternative scaffolds have been identified from natu-
ral product sources and are of particular interest here due to
their demonstrated antimycobacterial activity. For example N-
methyl-2-alkenyl-4-quinolone derivatives isolated from Evodia
rutaecarpaand aporphine alkaloids isolated from Ocotea macro-
phylla inhibit the activity of MurE from M. tuberculosis (9, Fig. 6,)
and prevent cell growth at micromolar concentrations (Guz-
man et al. 2010; Guzman et al. 2011; Wube et al. 2011; Guzman
et al. 2013b). These natural products also inhibit the growth of
a number of other Gram-positive and Gram-negative bacteria,
but at much higher concentrations than those required against
mycobacteria. It has yet to be demonstrated that the observed
antibacterial activity is due to inhibition of MurE only, however,
given the relatively low affinity and lipophilicity of these com-
pounds it is unlikely that they are highly selective for their tar-
get. The natural product S-leucoxine from Rhodostemonodaphne
crenaticupula and related synthetic 1-tetrahydroquinolines have
also been demonstrated to inhibit MurE and mycobacterial cell
growth, further investigation, however suggested a pleiotropic
mechanism of action (Pesnot et al. 2011; Guzman et al. 2015).
A significant number of potent nanomolar inhibitors have
been developed for MurF. Abbott Laboratories, for example,
developed a series of cyanothiophene ligands with micromolar
to nanomolar affinity (10, Fig. 6) (Gu et al. 2004; Stamper et al.
2006) and Johnson & Johnson developed a series of potent thi-
azolylaminopyrimidine containing inhibitors (11, Fig. 6) (Baum
et al. 2006). However, none of the compounds tested demon-
strated any significant antibacterial activity. Finally, the identifi-
cation of diarylquinolines as low micromolar inhibitors of MurF
led to the development of 12 (Fig. 6), which exhibited an MIC
of 8 μg/mL against a number of bacterial species (Enterococcus
faecalis, E. faecium and S. aureus). Compound 12 binds MurF in
the presence of ATP and so is likely to occupy the substrate
binding site. In addition, treatment of LPS deficient E. coli with
12 resulted in the accumulation of the substrate of MurF, how-
ever, over expression of MurF did not significantly affect the
MIC. Hence, although the diarlyquinolines represent a promis-
ing start point for the development of novel antibacterial agents,
the mechanism of action warrants further examination.
Strategies for the development of multi-target inhibitors of
the Mur ligases fall in to two categories: (i) ligands that mimic
the shared tetrahedral intermediate of the Mur ligases and
(ii) ligands designed to target the ATP-binding site, which has
the greatest sequence and structural homology amongst the
enzymes. Alternatively, for dual target inhibitors, the substrate
binding sites of MurC/D and MurE/F are likely to have sufficient
similarity to allow the design of dual inhibitors. The majority
of efforts to date have focussed on the first strategy leading to
benzene-1,3-dicarboxylic acid 2,5-dimethylpyrrole derivatives
(Fig. 6, 13 & 14) (Perdih et al. 2014) and napthyl tetronic acids
(Mansour et al. 2007) as first multi-target lead compounds. To
evaluate the potential of these Mur ligase inhibitors as inhibitors
of the mycobacterial enzymes, we performed a docking study
(Autodock Vina) (Trott and Olson 2010), using a select number
of inhibitors and MurE (pdb code: 2XJA) from M. tuberculosis as a
model system. The pyrazalopyrimidine 5 was designed to inhibit
MurC from E. coli (Deva et al. 2006). Docking of 5 to MurE from
M. tuberculosis resulted in the prediction of two major binding
modes. In the first binding mode there is good overlap between
the pyrazolpyrimidine ring and the adenosine ring of ATP (as
observed X-ray crystal structure) allowing a π-π stacking inter-
action with Tyr343 (an aromatic residue at this position is highly
conserved amongst the Mur ligases); the pyrazole ring is pre-
dicted to occupy the phosphate binding site and forms a H-
bonding interaction with Thr158 (part of the conserved diphos-
phate binding motif) (Fig. 7). In the second binding mode the
pyrazole ring forms a π-π stacking interaction with Tyr343, the
pyrazolopyrimidine moiety is predicted to form polar contacts
with two key residues (Asn347 and Thr158) and the hydroxyl
group interacts with residues in the highly conserved phosphate
binding loop (Gly156, Lys157, Thr158) (Fig. 7B). Given the reca-
pitulation of key interactions formed by ATP within the bind-
ing site and similarities with the binding mode predicted by
Hameed et al. it is feasible that 5 and analogues thereof may pro-
vide a starting points for the development of inhibitors for the
mycobacterial Mur ligases (Hameed P et al. 2014).
The benzylidene rhodanine inhibitor 7 was designed as an
inhibitor of MurD from E. coli (Zidar et al. 2010). In the X-ray
crystal structure 7 was found to span the substrate binding site
with the rhodanine heterocycle acting as a diphosphate mimic.
This binding mode is recapitulated when docked into MurE from
M. tuberculosis, however, because of the larger binding site the
 Maitra et al. 565

Figure 7. Representative predictive binding pockets for selected compounds to MurE (PDB code 2XJA) from M. tuberculosis (overlaid with ATP or UAG as appropriate,
shown in lines) (A), Compound 5, predicted binding affinity −9.2 kcal/mol (B), Compound 5, predicted binding affinity −8.9 kcal/mol (C), Compound 8, predicted binding
affinity -7.8 kcal/mol (D) Compound 8, predicted binding affinity −7.2 kcal/mol (E) Compound 14, predicted binding affinity −6.8 kcal/mol.

rhodanine heterocycle is predicted to bind in the hydrophobic
pocket occupied by the uracil moiety of UAG and not to the
diphosphate binding site as in E. coli MurD.
Transition state analogue, 8, another inhibitor of MurD,
which contains a conformationally restrained Glu analogue and
a sulphonamide moiety as a transition state analogue, was also
docked to MurE. This inhibitor was predicted to span the sub-
strate binding site and two alternate binding modes were pre-
dicted: (i) in which the Glu mimic occupied the uracil binding
site and the sulphonamide the diphosphate binding site and (ii)
in which the cyanobenzyl functional group occupied the uracil
binding site and the sulphonamide and Glu mimic mapped
well onto the observed conformation of UDP-MurNac-peptide
in the X-ray crystal structure of MurE (Fig. 7C, D). However, it
is unlikely that the sulphonamide moiety will be able to bind
in a similar conformation to the tetrahedral transition state due
to the larger and more open binding site of MurE in compari-
sion to MurD. It is important to note that the N-terminal domain
exhibits the greatest sequence and structural variation i.e. the
N-terminal domain in MurC/D is significantly different to that in
MurE/F. This domain is also more flexible and undergoes signifi-
cant conformational changes on ligand binding; this flexibility is
required to enable the Mur ligase to accommodate the growing
peptide chain of the PG precursor (Basavannacharya et al. 2010b).
 566 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
As docking experiments are based solely on and restricted by the
structures/conformations available, these need to be integrated
with in vitro experiments such as enzyme inhibition kinetics to
gain confidence in the targeting of inhibitors.
A related set of compounds designed as multi-target Mur lig-
ase inhibitors are compounds 13 (a dual inhibitor of MurD/E)
and 14 (an inhibitor of MurC-F). These were docked against MurE
from M. tuberculosis (Tomasic et al. 2010; Basavannacharya et al.
2010a; Perdih et al. 2014). For 13, as expected, the dihydopyrim-
idine moiety was predicted to act as a diphosphate mimic and
the phenyl ring to occupy the uracil binding site. The remain-
der of the inhibitor maps well onto the conformation of UAG
observed in the X-ray crystal structure ( 13 is, however, trun-
cated compared to UAG and so does not span the entire binding
site). 14, which is a moderate inhibitor of MurC-F, is predicted to
adopt a similar conformation within the substrate binding site
with the rhodanine heterocycle again acting as a diphosphate
mimic and the chlorophenyl moiety occupying the uracil bind-
ing site. The increased flexibility in 14, compared to its precur-
sor 13, allows the heterocycle to form polar contacts with key
residues in the diphosphate binding loop (Gly83, Ser84). Interac-
tions with key residues in the substrate binding site are picked
up by the isophthalic acid moiety (Arg230, Ser222, Thr195 and
Thr196) (Fig. 7E). The recapitulation of key interactions between
substrate and protein supports the conclusion that these lig-
ands likely bind to and inhibit mycobacterial MurE, although the
inherent flexibility of the Mur ligases is likely to complicate this
picture. Experimental evidence is needed to confirm inhibition
of mycobacterial Mur ligases by any of the inhibitors discussed
here.
As can be seen from this survey of the literature, much
progress towards the development of potent inhibitors of the
Mur ligases has been made. However, the translation from effec-
tive enzyme inhibition to antibacterial action is a significant
challenge. Membrane permeability and active efflux are likely
to present the biggest hurdles and many inhibitors developed to
date are very polar, particularly the phosphinate transition state
analogues and amino acid derivatives, making them unlikely
to easily traverse the membrane (Jarlier and Nikaido 1994). For
other classes of inhibitors, the explanation is less clear.
It is also important to highlight that for those compounds
with a demonstrated antibacterial activity there is often no evi-
dence for the observed bactericidal effect being caused by inhibi-
tion of a Mur ligase (Silver 2011). The IC50 /MIC data in conjunc-
tion with molecular modelling or X-ray crystallography, would
significantly inform the design of next generation of inhibitors.
The study by Barreteau et al. (Barreteau et al. 2012), which eval-
uated a selection of inhibitors designed to target MurD from E.
coli against orthologues from different bacteria is a step in this
direction.
Many of the inhibitors discussed contain what would typ-
ically be considered undesirable functional groups, for exam-
ple rhodanine and thiazolidindione moieties are notorious
pan assay interference compounds (PAINS) (Baell 2010; Baell
and Holloway 2010). These compounds often prove to be not
amenable to optimisation, preventing the progression of hit/lead
molecules. The pitiful success of HTS as a source of novel
antibacterial agents has been the subject of a comprehen-
sive commentary in the literature (Tomašić and Mašič 2012).
Hits derived from HTS campaigns are typically quite lipophilic,
whereas antibiotics are typically larger and more polar than
drugs designed for human targets (Payne et al. 2007). It is, there-
fore, questionable whether the libraries amassed by the phar-
maceutical industry are optimal for antibacterial drug discovery.
Fragment-based drug discovery may offer a viable alternative
(Hajduk and Greer 2007). This approach employs libraries of low
molecular weight compounds or fragments, these could include
commercial fragment libraries, repurposed kinase focussed
libraries to target the ATP binding site and libraries of phos-
phate bioisosteres (Huang et al. 2009; Volkamer et al. 2015). The
fragments are screened against the protein targets of interest
using a range of biophysical assays which have a much lower
false positive hit rate compared to the biochemical assays typ-
ically used in HTS screens (Schulz and Hubbard 2009). One of
the major advantages of this approach, and which is partic-
ularly relevant to antimicrobial drug discovery, is that due to
the nature of the hits identified, i.e. low molecular weight, low-
affinity and high-ligand efficiency, one can exert greater con-
trol over the physiochemical properties of the inhibitor dur-
ing the process of hit to lead development (Hajduk 2006). The
process of fragment optimisation is typically design intensive
and structure guided, however, given the current lack of struc-
tural information for the mycobacterial enzymes, in situ opti-
misation using kinetic target guided synthesis has great poten-
tial and would allow rapid exploration of the protein bind-
ing landscape and the inherent flexibility of the Mur ligases
(Hajduk 2006).
It is clear from the progress to date that to obtain anti-
tubercular agents that target the Mur ligases a number of hur-
dles need to be overcome. As a result, further exploration of
the Mur ligases as drug targets and the mechanisms of the
inhibitors developed to date are necessary if therapeutically use-
ful molecules are to be developed.
CONCLUSIONS AND FUTURE OUTLOOK
Due to the protective role of PG and its absence in humans,
its biosynthetic pathway remains an excellent therapeutic tar-
get (Strancar et al. 2007; Zawadzke et al. 2008; Tomasic et al.
2010). Encouragingly, technologies for high throughput screen-
ing against these targets have advanced (Kristan et al. 2009; Turk
et al. 2009; Guzman et al. 2013a). We would hope that these devel-
opments and the new insights into the molecular mechanisms
of cell wall synthesis could lead to the development of targeted
anti-mycobacterials.
SUPPLEMENTARY DATA
Supplementary data are available at FEMSRE online
ACKNOWLEDGEMENTS
We are grateful to Professor Simon Gibbons and Professor Tim-
othy D. McHugh for their helpful suggestions. SB would like
to thank MRC, UK for a New Investigators Research Grant
(GO801956). AM and LTM would like to thank the Wellcome Trust
(108877/Z/15/Z) for their PhD studentship. WV received support
from the BBSRC (BB/P001289/1).
The authors acknowledge that the work of many eminent
researchers could not be cited in the review due to limited
space. We are grateful for their collective research contribut-
ing to advances in technology and our understanding of the
mycobacterial cell envelope.
Conflict of interest. None declared.
 Maitra et al. 567

REFERENCES Becker K, Sander P. Mycobacterium tuberculosis lipoproteins in vir-

ulence and immunity–fighting with a double-edged sword.
Ahangar MS, Furze CM, Guy CS et al. Structural and functional
FEBS Lett 2016;590:3800–19.
determination of homologs of the Mycobacterium tuberculosis
Benson TE, Marquardt JL, Marquardt AC et al. Overexpres-
N-acetylglucosamine-6-phosphate deacetylase (NagA). J Biol
sion, purification, and mechanistic study of UDP-N-
Chem 2018;293:9770–83.
acetylenolpyruvylglucosamine reductase. Biochemistry
Al-Dabbagh B, Henry X, Ghachi ME et al. Active Site Map-
1993;32:2024–30.
ping of MraY, a Member of the Polyprenyl-phosphate N-
Bergfeld AK, Pearce OM, Diaz SL et al. Metabolism of ver-
Acetylhexosamine 1-Phosphate Transferase Superfamily,
tebrate amino sugars with N-glycolyl groups: elucidating
Catalyzing the First Membrane Step of Peptidoglycan Biosyn-
the intracellular fate of the non-human sialic acid N-
thesis†. Biochemistry 2008;47:8919–28.
glycolylneuraminic acid. J Biol Chem 2012;287:28865–81.
Aldridge BB, Fernandez-Suarez M, Heller D et al. Asymmetry
Bertrand JA, Auger G, Martin L et al. Determination of the MurD
and aging of mycobacterial cells lead to variable growth and
mechanism through crystallographic analysis of enzyme
antibiotic susceptibility. Science 2012;335:100–4.
complexes. J Mol Biol 1999;289:579–90.
Anderson MS, Eveland SS, Onishi HR et al. Kinetic mecha-
Bhakta S, Basu J. Overexpression, purification and biochemical
nism of the Escherichia coli UDPMurNAc-tripeptide D-alanyl-
characterization of a class A high-molecular-mass penicillin-
D-alanine-adding enzyme: use of a glutathione S-transferase
binding protein (PBP), PBP1∗ and its soluble derivative from
fusion. Biochemistry 1996;35:16264–9.
Mycobacterium tuberculosis. Biochem J 2002;361:635–9.
Anuradha CM, Mulakayala C, Babajan B et al. Probing ligand
Bhattacharjee N, Triboulet S, Dubée V et al. Negative impact
binding modes of Mycobacterium tuberculosis MurC ligase by
of carbapenem methylation on the reactivity of β-lactams
molecular modeling, dynamics simulation and docking. J Mol
for cysteine acylation revealed by quantum calculations and
Model 2010;16:77–85.
kinetic analyses. Antimicrob Agents Chemother 2019; AAC.
Arvind A, Kumar V, Saravanan P et al. Homology modeling,
e02039-18.
molecular dynamics and inhibitor binding study on MurD
Bickford JS, Nick HS. Conservation of the PTEN catalytic motif
ligase of Mycobacterium tuberculosis. Interdiscip Sci, Comput Life
in the bacterial undecaprenyl pyrophosphate phosphatase,
Sci 2012;4:223–38.
BacA/UppP. Microbiology 2013;159:2444–55.
Awasthy D, Bharath S, Subbulakshmi V et al. Alanine racemase
Bielnicki J, Devedjiev Y, Derewenda U et al. B. subtilis ykuD pro-
mutants of Mycobacterium tuberculosis require D-alanine for
tein at 2.0 A resolution: insights into the structure and func-
growth and are defective for survival in macrophages and
tion of a novel, ubiquitous family of bacterial enzymes. Pro-
mice. Microbiology 2012;158:319–27.
teins 2006;62:144–51.
Babajan B, Chaitanya M, Rajsekhar C et al. Comprehensive struc-
Bolla JR, Sauer JB, Wu D et al. Direct observation of the influ-
tural and functional characterization of Mycobacterium tuber-
ence of cardiolipin and antibiotics on lipid II binding to MurJ.
culosis UDP-NAG enolpyruvyl transferase (Mtb-MurA) and
Nature Chem 2018;10:363.
prediction of its accurate binding affinities with inhibitors.
Botella H, Vaubourgeix J, Lee MH et al. Mycobacterium tuberculo-
Interdiscip Sci, Comput Life Sci 2011;3:204–16.
sis protease MarP activates a peptidoglycan hydrolase during
Baell JB, Holloway GA. New substructure filters for removal of
acid stress. EMBO J 2017a;36:536–48.
pan assay interference compounds (PAINS) from screening
Botella H, Yang G, Ouerfelli O et al. Distinct spatiotemporal
libraries and for their exclusion in bioassays. J Med Chem
dynamics of peptidoglycan synthesis between Mycobac-
2010;53:2719–40.
terium smegmatis and Mycobacterium tuberculosis. MBio
Baell JB. Observations on screening-based research and some
2017b;8:e01183–17.
concerning trends in the literature. Future Med Chem
Both D, Schneider G, Schnell R. Peptidoglycan remodeling in
2010;2:1529–46.
Mycobacterium tuberculosis: comparison of structures and cat-
Baranowski C, Welsh MA, Sham L-T et al. Maturing Mycobacterium
alytic activities of RipA and RipB. J Mol Biol 2011;413:247–60.
smegmatis peptidoglycan requires non-canonical crosslinks
Both D, Steiner EM, Stadler D et al. Structure of LdtMt2, an L,D-
to maintain shape. eLife 2018;7:e37516.
transpeptidase from Mycobacterium tuberculosis. Acta Crystal-
Barreteau H, Sosic I, Turk S et al. MurD enzymes from dif-
logr Sec D, Biol crystallogr 2013;69: 432–41.
ferent bacteria: evaluation of inhibitors. Biochem Pharmacol
Boudreau MA, Fisher JF, Mobashery S. Messenger functions of
2012;84:625–32.
the bacterial cell wall-derived muropeptides. Biochemistry
Basavannacharya C, Moody PR, Munshi T et al. Essential
2012;51:2974–90.
residues for the enzyme activity of ATP-dependent MurE
Bourai N, Jacobs WR, Jr., Narayanan S. Deletion and overex-
ligase from Mycobacterium tuberculosis. Protein Cell 2010b;1:
pression studies on DacB2, a putative low molecular mass
1011–22.
penicillin binding protein from Mycobacterium tuberculosis
Basavannacharya C, Robertson G, Munshi T et al. ATP-dependent
H(37)Rv. Microb Pathog 2012;52: 109–16.
MurE ligase in Mycobacterium tuberculosis: biochemical and
Boutte CC, Baer CE, Papavinasasundaram K et al. A cytoplasmic
structural characterisation. Tuberculosis (Edinb) 2010a;90:16–
peptidoglycan amidase homologue controls mycobacterial
24.
cell wall synthesis. eLife 2016;5:e14590.
Baum EZ, Crespo-Carbone SM, Abbanat D et al. Utility of
Brennan PJ, Nikaido H. The envelope of mycobacteria. Annu Rev
muropeptide ligase for identification of inhibitors of the cell
Biochem 1995;64:29–63.
wall biosynthesis enzyme MurF. Antimicrob Agents Chemother
Brennan PJ. Structure, function, and biogenesis of the cell wall of
2006;50:230–6.
Mycobacterium tuberculosis. Tuberculosis (Edinb) 2003;83:91–7.
Baum EZ, Montenegro DA, Licata Let al. Identification and
Bronson JJ, DenBleyker KL, Falk PJet al. Discovery of the first
characterization of new inhibitors of the Escherichia coli
antibacterial small molecule inhibitors of MurB. Bioorg. Med.
MurA enzyme. Antimicrob. Agents Chemother 2001; 45:3182–
Chem. Lett. 2003;13 :873–5. 12617911
8. 11600375
 568 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
Brown ED, Vivas EI, Walsh CT et al. MurA (MurZ), the enzyme that
catalyzes the first committed step in peptidoglycan biosyn-
thesis, is essential in Escherichia coli. J Bacteriol 1995;177: 4194–
7.
Bruning JB, Murillo AC, Chacon O et al. Structure of the Mycobac-
terium tuberculosis D-alanine:D-alanine ligase, a target of
the antituberculosis drug D-cycloserine. Antimicrob Agents
Chemother 2011;55:291–301.
Carel C, Nukdee K, Cantaloube S et al. Mycobacterium tuberculo-
sis proteins involved in mycolic acid synthesis and transport
localize dynamically to the old growing pole and septum.
PLoS One 2014;9:e97148.
Cava F, de Pedro MA, Lam H et al. Distinct pathways for modi-
fication of the bacterial cell wall by non-canonical D-amino
acids. EMBO J 2011;30:3442–53.
Chalker AF, Ingraham KA, Lunsford RD et al. The bacA gene,
which determines bacitracin susceptibility in Streptococcus
pneumoniae and Staphylococcus aureus, is also required for vir-
ulence. Microbiology 2000;146:1547–53.
Chaloupka J, Strnadova M. Turnover of murein in a
diaminopimelic acid dependent mutant of Escherichia
coli. Folia Microbiol (Praha) 1972;17:446–55.
Chamaillard M, Girardin SE, Viala J et al. Nods, Nalps and Naip:
intracellular regulators of bacterial-induced inflammation.
Cell Microbiol 2003a;5:581–92.
Chamaillard M, Hashimoto M, Horie Y et al. An essential role for
NOD1 in host recognition of bacterial peptidoglycan contain-
ing diaminopimelic acid. Nat Immunol 2003b;4:702–7.
Chambers HF, Moreau D, Yajko D et al. Can penicillins and other
beta-lactam antibiotics be used to treat tuberculosis? Antimi-
crob Agents Chemother 1995;39:2620–4.
Chao MC, Kieser KJ, Minami S et al. Protein complexes and prote-
olytic activation of the cell wall hydrolase RipA regulate sep-
tal resolution in mycobacteria. PLoS Pathog 2013;9:e1003197.
Chauviac FX, Robertson G, Quay DH et al. The RpfC (Rv1884)
atomic structure shows high structural conservation within
the resuscitation-promoting factor catalytic domain. Acta
Crystallogr F Struct Biol Commun 2014;70:1022–6.
Cheng Q, Park JT. Substrate specificity of the AmpG permease
required for recycling of cell wall anhydro-muropeptides. J
Bacteriol 2002;184:6434–6.
Chesnokova ON, McPherson SA, Steichen CT et al. The spore-
specific alanine racemase of Bacillus anthracis and its role in
suppressing germination during spore development. J Bacte-
riol 2009;191:1303–10.
Chung BC, Zhao J, Gillespie RA et al. Crystal structure of MraY, an
essential membrane enzyme for bacterial cell wall synthesis.
Science 2013;341:1012–6.
Cohen-Gonsaud M, Barthe P, Bagneris C et al. The structure of a
resuscitation-promoting factor domain from Mycobacterium
tuberculosis shows homology to lysozymes. Nat Struct Mol Biol
2005;12:270–3.
Cohen-Gonsaud M, Keep NH, Davies AP et al. Resuscitation-
promoting factors possess a lysozyme-like domain. Trends
Biochem Sci 2004;29:7–10.
Cole ST, Brosch R, Parkhill J et al. Deciphering the biology
of Mycobacterium tuberculosis from the complete genome
sequence. Nature 1998;393:537–44.
Craggs PD, Mouilleron S, Rejzek M et al. The mechanism of acetyl
transfer catalyzed by Mycobacteri um tuberculosis GlmU. Bio-
chemistry 2018;57:3387–401.
Crick DC, Mahapatra S, Brennan PJ. Biosynthesis of the
arabinogalactan-peptidoglycan complex of Mycobacterium
tuberculosis. Glycobiology 2001;11:107R–18R.
Crick DC, Schulbach MC, Zink EE et al. Polyprenyl phosphate
biosynthesis in Mycobacterium tuberculosis and Mycobacterium
smegmatis. J Bacteriol 2000;182:5771–8.
Dajkovic A, Tesson B, Chauhan S et al. Hydrolysis of peptidogly-
can is modulated by amidation of meso-diaminopimelic acid
and Mg(2+) in Bacillus subtilis. Mol Microbiol 2017;104:972–88.
Daniel RA, Errington J. Control of cell morphogenesis in bac-
teria: two distinct ways to make a rod-shaped cell. Cell
2003;113:767–76.
Darby CM, Venugopal A, Ehrt S et al. Mycobacterium tuberculosis
gene Rv2136c is dispensable for acid resistance and virulence
in mice. Tuberculosis 2011;91:343–7.
Dasgupta A, Datta P, Kundu M et al. The serine/threonine kinase
PknB of Mycobacterium tuberculosis phosphorylates PBPA, a
penicillin-binding protein required for cell division. Microbi-
ology 2006;152:493–504.
David S. Synergic activity of d-cycloserine and β-chloro-
d-alanine against Mycobacterium tuberculosis. J Antimicrob
Chemother 2001;47:203–6.
DeJesus MA, Gerrick ER, Xu W et al. Comprehensive essentiality
analysis of the Mycobacterium tuberculosis genome via satu-
rating transposon mutagenesis. MBio 2017;8:e02133–16.
Dementin S. Etude du mécanisme réactionnel des Mur
synthétases, enzymes impliquées dans la biosynthèse
du peptidoglycane bactérien. 2001.
den Blaauwen T, de Pedro MA, Nguyen-Disteche M et al.
Morphogenesis of rod-shaped sacculi. FEMS Microbiol Rev
2008;32:321–44.
De Smet KA, Kempsell KE, Gallagher A et al. Alteration of a
single amino acid residue reverses fosfomycin resistance of
recombinant MurA from Mycobacterium tuberculosis. Microbi-
ology 1999;145:3177–84.
Deva T, Baker EN, Squire CJ et al. Structure of Escherichia coli
UDP-N-acetylmuramoyl:L-alanine ligase (MurC). Acta crystal-
lographica Section D, Biological crystallography 2006;62:1466–74.
Dmitriev B, Toukach F, Ehlers S. Towards a comprehensive view
of the bacterial cell wall. Trends Microbiol 2005;13:569–74.
Dmitriev BA, Toukach FV, Schaper KJ et al. Tertiary struc-
ture of bacterial murein: the scaffold model. J Bacteriol
2003;185:3458–68.
Dominguez-Escobar J, Chastanet A, Crevenna AH et al. Proces-
sive movement of MreB-associated cell wall biosynthetic
complexes in bacteria. Science 2011;333:225–8.
Domı́nguez-Escobar J, Chastanet A, Crevenna AH et al. Proces-
sive movement of MreB-associated cell wall biosynthetic
complexes in bacteria. Science 2011;333:225–8.
Downing KJ, Mischenko VV, Shleeva MO et al. Mutants of
Mycobacterium tuberculosis lacking three of the five rpf-like
genes are defective for growth in vivo and for resuscitation
in vitro. Infect Immun 2005;73:3038–43.
Dubee V, Triboulet S, Mainardi JL et al. Inactivation of Mycobac-
terium tuberculosis L,D-Transpeptidase Ldt(Mt1) by Car-
bapenems and Cephalosporins. Antimicrob Agents Chemother
2012;56:4189–95.
Du W, Brown JR, Sylvester DR et al. Two active forms of UDP-N-
acetylglucosamine enolpyruvyl transferase in gram-positive
bacteria. J Bacteriol 2000;182:4146–52.
Dziadek B, Brzostek A, Grzybowski M et al. Mycobacterium tuber-
culosis AtsG (Rv0296c), GlmU (Rv1018c) and SahH (Rv3248c)
Proteins Function as the Human IL-8-Binding Effectors and
Contribute to Pathogen Entry into Human Neutrophils. PLoS
One 2016;11:e0148030.
El Ghachi M, Derbise A, Bouhss A et al. Identification of multi-
ple genes encoding membrane proteins with undecaprenyl

pyrophosphate phosphatase (UppP) activity in Escherichia coli.
J Biol Chem 2005;280:18689–95.
El Zoeiby A, Sanschagrin F, Darveau A et al. Identification of
novel inhibitors of Pseudomonas aeruginosa MurC enzyme
derived from phage-displayed peptide libraries. J Antimicrob
Chemother 2003;51:531–43.
Emami K, Guyet A, Kawai Y et al. RodA as the missing gly-
cosyltransferase in Bacillus subtilis and antibiotic discov-
ery for the peptidoglycan polymerase pathway. Nat Microbiol
2017;2:16253.
Eniyan K, Dharavath S, Vijayan R et al. Crystal structure of UDP-
N-acetylglucosamine-enolpyruvate reductase (MurB) from
Mycobacterium tuberculosis. Biochimica et biophysica acta Proteins
and proteomics 2018;1866:397–406.
Eniyan K, Kumar A, Rayasam GV et al. Development of a one-
pot assay for screening and identification of Mur pathway
inhibitors in Mycobacterium tuberculosis. Sci Rep 2016;6:35134.
Erdemli SB, Gupta R, Bishai WR et al. Targeting the cell wall of
Mycobacterium tuberculosis: structure and mechanism of L,D-
transpeptidase 2. Structure 2012;20:2103–15.
Errington J. L-form bacteria, cell walls and the origins of life. Open
biology 2013;3:120143.
Esin S, Counoupas C, Aulicino A et al. Interaction of Mycobac-
terium tuberculosis cell wall components with the human nat-
ural killer cell receptors NKp44 and Toll-like receptor 2. Scand
J Immunol 2013;77:460–9.
Eveland SS, Pompliano DL, Anderson MS. Conditionally lethal
Escherichia coli murein mutants contain point defects that
map to regions conserved among murein and folyl poly-
gamma-glutamate ligases: identification of a ligase super-
family. Biochemistry 1997;36:6223–9.
Falk PJ, Ervin KM, Volk KS et al. Biochemical evidence for the
formation of a covalent acyl-phosphate linkage between
UDP-N-acetylmuramate and ATP in the Escherichia coli UDP-
N-acetylmuramate:L-alanine ligase-catalyzed reaction. Bio-
chemistry 1996;35:1417–22.
Feng Z, Barletta RG. Roles of Mycobacterium smegmatis D-
alanine:D-alanine ligase and D-alanine racemase in the
mechanisms of action of and resistance to the peptido-
glycan inhibitor D-cycloserine. Antimicrob Agents Chemother
2003;47:283–91.
Figueiredo TA, Sobral RG, Ludovice AM et al. Identification of
genetic determinants and enzymes involved with the amida-
tion of glutamic acid residues in the peptidoglycan of Staphy-
lococcus aureus. PLoS Pathog 2012;8:e1002508.
Filippova EV, Kieser KJ, Luan CH et al. Crystal structures
of the transpeptidase domain of the Mycobacterium tuber-
culosis penicillin-binding protein PonA1 reveal potential
mechanisms of antibiotic resistance. FEBS J 2016;283:
2206–18.
Flores AR, Parsons LM, Pavelka MS, Jr. Characterization of
novel Mycobacterium tuberculosis and Mycobacterium smegma-
tis mutants hypersusceptible to beta-lactam antibiotics. J
Bacteriol 2005;187:1892–900.
Forrellad MA, Klepp LI, Gioffré A et al. Virulence factors of the
Mycobacterium tuberculosis complex. Virulence 2013;4:3–66.
Garcia-Heredia A, Pohane AA, Melzer ES et al. Peptidogly-
can precursor synthesis along the sidewall of pole-growing
mycobacteria. eLife 2018;7:e37243.
Garner EC, Bernard R, Wang W et al. Coupled, circumferential
motions of the cell wall synthesis machinery and MreB fila-
ments in B. subtilis. Science 2011;333:222–5.
Ghuysen JM. Serine beta-lactamases and penicillin-binding pro-
teins. Annu Rev Microbiol 1991;45:37–67.
Maitra et al. 569
Gilleron M, Quesniaux VF, Puzo G. Acylation state of the
phosphatidylinositol hexamannosides from Mycobacterium
bovis bacillus Calmette Guerin and Mycobacterium tuberculosis
H37Rv and its implication in Toll-like receptor response. J Biol
Chem 2003;278:29880–9.
Girardin SE, Boneca IG, Viala J et al. Nod2 is a general sensor of
peptidoglycan through muramyl dipeptide (MDP) detection.
J Biol Chem 2003a;278:8869–72.
Girardin SE, Travassos LH, Herve M et al. Peptidoglycan molecu-
lar requirements allowing detection by Nod1 and Nod2. J Biol
Chem 2003b;278:41702–8.
Glauner B, Holtje JV, Schwarz U. The composition of the murein
of Escherichia coli. J Biol Chem 1988;263:10088–95.
Goffin C, Ghuysen JM. Multimodular penicillin-binding proteins:
an enigmatic family of orthologs and paralogs. Microbiol Mol
Biol Rev 1998;62:1079–93.
Goodell EW, Schwarz U. Release of cell wall peptides into culture
medium by exponentially growing Escherichia coli. J Bacteriol
1985;162:391–7.
Goodell EW. Recycling of murein by Escherichia coli. J Bacteriol
1985;163:305–10.
Griffin JE, Gawronski JD, DeJesus MA et al. High-resolution
phenotypic profiling defines genes essential for mycobac-
terial growth and cholesterol catabolism. PLoS Pathog
2011;7:e1002251.
Grzegorzewicz AE, De Sousa-d’Auria C, McNeil MR et al. Assem-
bling of the Mycobacterium tuberculosis cell wall core. J Biol
Chem 2016;291:18867–79.
Gupta R, Lavollay M, Mainardi JL et al. The Mycobacterium
tuberculosis protein LdtMt2 is a nonclassical transpeptidase
required for virulence and resistance to amoxicillin. Nat Med
2010;16:466–9.
Gu YG, Florjancic AS, Clark RF et al. Structure-activity relation-
ships of novel potent MurF inhibitors. Bioorganic & medicinal
chemistry letters 2004;14:267–70.
Guzman JD, Evangelopoulos D, Gupta A et al. Antimycobacteri-
als from lovage root (Ligusticum officinale Koch). Phytother-
apy research: PTR 2013b;27:993–8.
Guzman JD, Evangelopoulos D, Gupta A et al. Antitubercular spe-
cific activity of ibuprofen and the other 2-arylpropanoic acids
using the HT-SPOTi whole-cell phenotypic assay. BMJ open
2013a;3:e002672.
Guzman JD, Gupta A, Evangelopoulos D et al. Anti-tubercular
screening of natural products from Colombian plants:
3-methoxynordomesticine, an inhibitor of MurE lig-
ase of Mycobacterium tuberculosis. J Antimicrob Chemother
2010;65:2101–7.
Guzman JD, Pesnot T, Barrera DA et al. Tetrahydroisoquinolines
affect the whole-cell phenotype of Mycobacterium tuberculo-
sis by inhibiting the ATP-dependent MurE ligase. J Antimicrob
Chemother 2015;70:1691–703.
Guzman JD, Wube A, Evangelopoulos D et al. Interaction of N-
methyl-2-alkenyl-4-quinolones with ATP-dependent MurE
ligase of Mycobacterium tuberculosis: antibacterial activity,
molecular docking and inhibition kinetics. J Antimicrob
Chemother 2011;66:1766–72.
Hajduk PJ, Greer J. A decade of fragment-based drug design:
strategic advances and lessons learned. Nat Rev Drug Discov-
ery 2007;6:211.
Hajduk PJ. Fragment-based drug design: how big is too big? J Med
Chem 2006;49:6972–6.
Hameed P S, Manjrekar P, Chinnapattu M et al. Pyrazolopyrim-
idines establish MurC as a vulnerable target in Pseudomonas
aeruginosa and Escherichia coli. ACS Chem Biol 2014;9:2274–82.
 570 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
Hansen JM, Golchin SA, Veyrier FJ et al. N-glycolylated pep-
tidoglycan contributes to the immunogenicity but not
pathogenicity of Mycobacterium tuberculosis. J Infect Dis
2014;209:1045–54.
Harth G, Masleša-Galić S, Tullius MV et al. All four Mycobac-
terium tuberculosis glnA genes encode glutamine synthetase
activities but only GlnA1 is abundantly expressed and
essential for bacterial homeostasis. Mol Microbiol 2005;58:
1157–72.
Hayashi JM, Richardson K, Melzer ES et al. Stress-induced reor-
ganization of the mycobacterial membrane domain. MBio
2018;9:e01823-17.
Heidrich C, Templin MF, Ursinus A et al. Involvement of N-
acetylmuramyl-L-alanine amidases in cell separation and
antibiotic-induced autolysis of Escherichia coli. Mol Microbiol
2001;41:167–78.
Heidrich C, Ursinus A, Berger J et al. Effects of multiple deletions
of murein hydrolases on viability, septum cleavage, and sen-
sitivity to large toxic molecules in Escherichia coli. J Bacteriol
2002;184:6093–9.
Hernández-Rocamora VM, Otten CF, Radkov A et al. Coupling of
polymerase and carrier lipid phosphatase prevents product
inhibition in peptidoglycan synthesis. Cell Surface 2018;2:1–
13.
Herve M, Boniface A, Gobec S et al. Biochemical character-
ization and physiological properties of Escherichia coli
UDP-N-acetylmuramate:L-alanyl-gamma-D-glutamyl-
meso-diaminopimelate ligase. J Bacteriol 2007;189: 3987–95.
Hett EC, Chao MC, Rubin EJ. Interaction and modulation of two
antagonistic cell wall enzymes of mycobacteria. PLoS Pathog
2010;6:e1001020.
Hett EC, Chao MC, Steyn AJ et al. A partner for the resuscitation-
promoting factors of Mycobacterium tuberculosis. Mol Microbiol
2007;66:658–68.
Hett EC, Rubin EJ. Bacterial growth and cell division: a mycobac-
terial perspective. Microbiol Mol Biol Rev 2008;72:126–56, table
of contents.
Ho H-T, Falk PJ, Ervin KM et al. UDP-N-acetylmuramyl-L-alanine
functions as an activator in the regulation of the Escherichia
coli glutamate racemase activity. Biochemistry 1995;34:2464–
70.
Holtje JV. Growth of the stress-bearing and shape-maintaining
murein sacculus of Escherichia coli. Microbiol Mol Biol Rev
1998;62:181–203.
Huang D, Zhou T, Lafleur K et al. Kinase selectivity potential for
inhibitors targeting the ATP binding site: a network analysis.
Bioinformatics 2009;26:198–204.
Hugonnet J-E, Blanchard JS. Irreversible inhibition of the
Mycobacterium tuberculosis β-lactamase by clavulanate. Bio-
chemistry 2007;46:11998–2004.
Hugonnet J-E, Mengin-Lecreulx D, Monton A et al. Factors
essential for L, D-transpeptidase-mediated peptidoglycan
cross-linking and β-lactam resistance in Escherichia coli. eLife
2016;5:e19469.
Hugonnet J-E, Tremblay LW, Boshoff HI et al. Meropenem-
clavulanate is effective against extensively drug-resistant
Mycobacterium tuberculosis. Science 2009;323:1215–8.
Humann J, Lenz LL. Bacterial peptidoglycan degrading enzymes
and their impact on host muropeptide detection. J Innate
Immun 2009;1:88–97.
Hussain S, Wivagg CN, Szwedziak P et al. MreB filaments align
along greatest principal membrane curvature to orient cell
wall synthesis. eLife 2018;7:e32471.
Inoue A, Murata Y, Takahashi H et al. Involvement of an essential
gene, mviN, in murein synthesis in Escherichia coli. J Bacteriol
2008;190:7298–301.
Jacobs C, Huang LJ, Bartowsky E et al. Bacterial cell wall recy-
cling provides cytosolic muropeptides as effectors for beta-
lactamase induction. EMBO J 1994;13:4684–94.
Jagtap PK, Soni V, Vithani N et al. Substrate-bound crystal
structures reveal features unique to Mycobacterium tuber-
culosis N-acetyl-glucosamine 1-phosphate uridyltransferase
and a catalytic mechanism for acetyl transfer. J Biol Chem
2012;287:39524–37.
Jankute M, Cox JA, Harrison J et al. Assembly of the mycobacterial
cell wall. Annu Rev Microbiol 2015;69:405–23.
Jarlier V, Nikaido H. Mycobacterial cell wall: structure and
role in natural resistance to antibiotics. FEMS Microbiol Lett
1994;123:11–8.
Johnson JW, Fisher JF, Mobashery S. Bacterial cell-wall recycling.
Ann N Y Acad Sci 2013;1277:54–75.
Joyce G, Williams KJ, Robb M et al. Cell division site place-
ment and asymmetric growth in mycobacteria. PLoS One
2012;7:e44582.
Kana BD, Gordhan BG, Downing KJ et al. The resuscitation-
promoting factors of Mycobacterium tuberculosis are required
for virulence and resuscitation from dormancy but are
collectively dispensable for growth in vitro. Mol Microbiol
2008;67:672–84.
Kastrinsky DB, Barry CE. Synthesis of labeled meropenem for the
analysis of M. tuberculosis transpeptidases. Tetrahedron Lett
2010;51: 197–200.
Kaur D, Brennan PJ, Crick DC. Decaprenyl diphosphate synthesis
in Mycobacterium tuberculosis. J Bacteriol 2004;186:7564–70.
Kieser KJ, Boutte CC, Kester JC et al. Phosphorylation of the pep-
tidoglycan synthase PonA1 governs the rate of polar elonga-
tion in mycobacteria. PLoS Pathog 2015;11:e1005010.
Kieser KJ, Rubin EJ. How sisters grow apart: mycobacterial
growth and division. Nat Rev Microbiol 2014;12:550.
Kilburn JO, Best GK. Characterization of autolysins from
Mycobacterium smegmatis. J Bacteriol 1977;129:750–5.
Killick KE, Ni Cheallaigh C, O’Farrelly C et al. Receptor-
mediated recognition of mycobacterial pathogens. Cell Micro-
biol 2013;15:1484–95.
Kim HS, Kim J, Im HN et al. Structural basis for the inhi-
bition of Mycobacterium tuberculosis L, D-transpeptidase by
meropenem, a drug effective against extensively drug-
resistant strains. Acta Crystallogr Sect D 2013;69:420–31.
Kluj RM, Ebner P, Adamek M et al. Recovery of the Peptidoglycan
Turnover Product released by the Autolysin Atl in Staphylo-
coccus aureus involves the Phosphotransferase System Trans-
porter MurP and the Novel 6-phospho-N-acetylmuramidase
MupG. Front Microbiol 2018;9.
Koch AL. Orientation of the peptidoglycan chains in the sacculus
of Escherichia coli. Res Microbiol 1998;149:689–701.
Koch AL. Simulation of the conformation of the murein fabric:
the oligoglycan, penta-muropeptide, and cross-linked nona-
muropeptide. Arch Microbiol 2000;174:429–39.
Korfmann G, Sanders CC. ampG is essential for high-level
expression of AmpC beta-lactamase in Enterobacter cloacae.
Antimicrob Agents Chemother 1989;33:1946–51.
Kotani S, Yanagida I, Kato K et al. Studies on peptides, glycopep-
tides and antigenic polysaccharide-glycopeptide complexes
isolated from an L-11 enzyme lysate of the cell walls of
Mycobacterium tuberculosis strain H37Rv. Biken J 1970;13:249–
75.

Koul A, Vranckx L, Dhar N et al. Delayed bactericidal response
of Mycobacterium tuberculosis to bedaquiline involves remod-
elling of bacterial metabolism. Nature communications
2014;5:3369.
Kristan K, Kotnik M, Oblak M et al. New high-throughput
fluorimetric assay for discovering inhibitors of UDP-N-
acetylmuramyl-L-alanine: D-glutamate (MurD) ligase. J
Biomol Screen 2009;14:412–8.
Kuk AC, Mashalidis EH, Lee SY. Crystal structure of the MOP flip-
pase MurJ in an inward-facing conformation. Nat Struct Mol
Biol 2017;24:171–6.
Kumar A, Kumar S, Kumar D et al. The structure of Rv3717 reveals
a novel amidase from Mycobacterium tuberculosis. Acta Crystal-
logr Sec D, Biol Crystallogr 2013;69:2543–54.
Kumar P, Arora K, Lloyd JR et al. Meropenem inhibits D, D-
carboxypeptidase activity in Mycobacterium tuberculosis. Mol
Microbiol 2012;86:367–81.
Kumar V, Saravanan P, Arvind A et al. Identification of hotspot
regions of MurB oxidoreductase enzyme using homology
modeling, molecular dynamics and molecular docking tech-
niques. J Mol Model 2011;17:939–53.
Lam H, Oh DC, Cava F et al. D-amino acids govern stationary
phase cell wall remodeling in bacteria. Science 2009;325:1552–
5.
Lavollay M, Arthur M, Fourgeaud M et al. The peptidoglycan of
stationary-phase Mycobacterium tuberculosis predominantly
contains cross-links generated by L,D-transpeptidation. J
Bacteriol 2008;190:4360–6.
Leclercq S, Derouaux A, Olatunji S et al. Interplay between
penicillin-binding proteins and SEDS proteins promotes bac-
terial cell wall synthesis. Sci Rep 2017;7:43306.
Lee TK, Huang KC. The role of hydrolases in bacterial cell-wall
growth. Curr Opin Microbiol 2013;16:760–6.
Legaree BA, Clarke AJ. Interaction of penicillin-binding protein
2 with soluble lytic transglycosylase B1 in Pseudomonas
aeruginosa. J Bacteriol 2008;190:6922–6.
Leisico F, D VV, Figueiredo TA et al. First insights of peptidoglycan
amidation in Gram-positive bacteria - the high-resolution
crystal structure of Staphylococcus aureus glutamine amido-
transferase GatD. Sci Rep 2018;8:5313.
LeMagueres P, Im H, Ebalunode J et al. The 1.9 A crystal structure
of alanine racemase from Mycobacterium tuberculosis con-
tains a conserved entryway into the active site. Biochemistry
2005;44:1471–81.
Levefaudes M, Patin D, De Sousa-D’Auria C et al. Diaminopimelic
acid amidation in Corynebacteriales: new insights into the
role of LtsA in peptidoglycan modification. J Biol Chem 2015:
jbc. M115. 642843.
Liechti G, Singh R, Rossi PL et al. Chlamydia trachomatis dapF
encodes a bifunctional enzyme capable of both d-glutamate
racemase and diaminopimelate epimerase activities. mBio
2018;9:e00204–18.
Liger D, Masson A, Blanot D et al. Study of the overpro-
duced uridine-diphosphate-N-acetylmuramate:L-alanine
ligase from Escherichia coli. Microb Drug Resist 1996;2:
25–7.
Li S, Kang J, Yu W et al. Identification of M. tuberculosis Rv3441c
and M. smegmatis MSMEG 1556 and essentiality of M. smeg-
matis MSMEG 1556. PLoS One 2012;7:e42769.
Li ZS, Beveridge TJ, Betts J et al. Partial characterization of
a major autolysin from Mycobacterium phlei. Microbiology
1999;145:169–76.
Maitra et al. 571
Loessner MJ, Kramer K, Ebel F et al. C-terminal domains of Lis-
teria monocytogenes bacteriophage murein hydrolases deter-
mine specific recognition and high-affinity binding to bacte-
rial cell wall carbohydrates. Mol Microbiol 2002;44:335–49.
Machowski EE, Senzani S, Ealand C et al. Comparative genomics
for mycobacterial peptidoglycan remodelling enzymes
reveals extensive genetic multiplicity. BMC Microbiol
2014;14:75.
Mahajan R. Bedaquiline: first FDA-approved tuberculosis drug in
40 years. Int J Appl Basic Med Res 2013;3:1–2.
Mahapatra S, Bhakta S, Ahamed J et al. Characterization of
derivatives of the high-molecular-mass penicillin-binding
protein (PBP) 1 of Mycobacterium leprae. Biochem J 2000;350 Pt
1:75–80.
Mahapatra S, Crick DC, McNeil MR et al. Unique structural fea-
tures of the peptidoglycan of Mycobacterium leprae. J Bacteriol
2008;190:655–61.
Mahapatra S, Scherman H, Brennan PJ et al. N Glycolylation of
the nucleotide precursors of peptidoglycan biosynthesis of
Mycobacterium spp. is altered by drug treatment. J Bacteriol
2005a;187:2341–7.
Mahapatra S, Yagi T, Belisle JT et al. Mycobacterial lipid II is
composed of a complex mixture of modified muramyl and
peptide moieties linked to decaprenyl phosphate. J Bacteriol
2005b;187:2747–57.
Manjunatha U, Boshoff HI, Barry CE. The mechanism of action
of PA-824: novel insights from transcriptional profiling. Com-
mun Integr Biol 2009;2:215–8.
Mansour TS, Caufield CE, Rasmussen B et al. Naphthyl tetronic
acids as multi-target inhibitors of bacterial peptidoglycan
biosynthesis. Chem Med Chem 2007;2:1414–7.
Marmor S, Petersen CP, Reck F et al. Biochemical characteriza-
tion of a phosphinate inhibitor of Escherichia coli MurC. Bio-
chemistry 2001;40:12207–14.
Marshall DD, Halouska S, Zinniel DK et al. Assessment of
metabolic changes in Mycobacterium smegmatis wild-type and
alr mutant strains: evidence of a new pathway of d-alanine
biosynthesis. J Proteome Res 2017;16:1270–9.
Mauck J, Chan L, Glaser L. Turnover of the cell wall of Gram-
positive bacteria. J Biol Chem 1971;246:1820–7.
Mavrici D, Prigozhin DM, Alber T. Mycobacterium tuberculosis RpfE
crystal structure reveals a positively charged catalytic cleft.
Protein Sci 2014;23:481–7.
Medzhitov R, Janeway CA. Innate immunity: the virtues of a non-
clonal system of recognition. Cell 1997;91:295–8.
Meeske AJ, Riley EP, Robins WP et al. SEDS proteins are a
widespread family of bacterial cell wall polymerases. Nature
2016;537:634.
Mengin-Lecreulx D, Falla T, Blanot D et al. Expression of
the Staphylococcus aureus UDP-N-acetylmuramoyl- L-alanyl-
D-glutamate:L-lysine ligase in Escherichia coli and effects
on peptidoglycan biosynthesis and cell growth. J Bacteriol
1999;181:5909–14.
Meniche X, Otten R, Siegrist MS et al. Subpolar addition of new
cell wall is directed by DivIVA in mycobacteria. Proc Natl Acad
Sci 2014;111:E3243–E51.
Meroueh SO, Bencze KZ, Hesek D et al. Three-dimensional struc-
ture of the bacterial cell wall peptidoglycan. Proc Natl Acad Sci
U S A 2006;103:4404–9.
Mikusova K, Slayden RA, Besra GS et al. Biogenesis of the
mycobacterial cell wall and the site of action of ethambutol.
Antimicrob Agents Chemother 1995;39:2484–9.
 572 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
Milligan DL, Tran SL, Strych U et al. The alanine racemase
of Mycobacterium smegmatis is essential for growth in the
absence of D-alanine. J Bacteriol 2007;189:8381–6.
Mingorance J, Tamames J, Vicente M. Genomic channeling in
bacterial cell division. J Mol Recognit 2004;17:481–7.
Mir M, Asong J, Li X et al. The extracytoplasmic domain of the
Mycobacterium tuberculosis Ser/Thr kinase PknB binds spe-
cific muropeptides and is required for PknB localization. PLoS
Pathog 2011;7:e1002182.
Mizyed S, Oddone A, Byczynski B et al. UDP-N-acetylmuramic
acid (UDP-MurNAc) is a potent inhibitor of MurA
(enolpyruvyl-UDP-GlcNAc synthase). Biochemistry
2005;44:4011–7.
Mohammadi T, Sijbrandi R, Lutters M et al. Specificity of the
transport of lipid II by FtsW in Escherichia coli. J Biol Chem
2014;289:14707–18.
Mohammadi T, van Dam V, Sijbrandi R et al. Identification of
FtsW as a transporter of lipid-linked cell wall precursors
across the membrane. EMBO J 2011;30: 1425–32.
Morayya S, Awasthy D, Yadav R et al. Revisiting the essential-
ity of glutamate racemase in Mycobacterium tuberculosis. Gene
2015;555:269–76.
Morlot C, Straume D, Peters K et al. Structure of the essential
peptidoglycan amidotransferase MurT/GatD complex from
Streptococcus pneumoniae. Nature Commun 2018;9:3180.
Morlot C, Uehara T, Marquis KA et al. A highly coordinated cell
wall degradation machine governs spore morphogenesis in
Bacillus subtilis. Genes Dev 2010;24:411–22.
Mortuza R, Aung HL, Taiaroa G et al. Overexpression of a newly
identified d-amino acid transaminase in Mycobacterium smeg-
matis complements glutamate racemase deletion. Mol Micro-
biol 2018;107:198–213.
Morè N, Martorana AM, Biboy J et al. Peptidoglycan remodel-
ing enables escherichia coli to survive severe outer membrane
assembly defect. mBio 2019;10:e02729–18.
Moynihan PJ, Cadby IT, Veerapen Net al. The hydrolase LpqI
primes mycobacterial peptidoglycan recycling. Nat Commun
2019; 10:2647.
Mukamolova GV, Kaprelyants AS, Young DI et al. A bacterial
cytokine. Proc Natl Acad Sci U S A 1998a;95:8916–21.
Mukamolova GV, Kormer SS, Kell DB et al. Stimulation of the
multiplication of Micrococcus luteus by an autocrine growth
factor. Arch Microbiol 1999;172:9–14.
Mukamolova GV, Murzin AG, Salina EG et al. Muralytic activity
of Micrococcus luteus Rpf and its relationship to physiological
activity in promoting bacterial growth and resuscitation. Mol
Microbiol 2006;59:84–98.
Mukamolova GV, Turapov OA, Kazarian K et al. The rpf gene of
Micrococcus luteus encodes an essential secreted growth fac-
tor. Mol Microbiol 2002;46:611–21.
Mukamolova GV, Yanopolskaya ND, Kell DB et al. On resuscita-
tion from the dormant state of Micrococcus luteus. Antonie Van
Leeuwenhoek 1998b;73:237–43.
Munch D, Roemer T, Lee SH et al. Identification and in
vitro analysis of the GatD/MurT enzyme-complex catalyz-
ing lipid II amidation in Staphylococcus aureus. PLoS Pathog
2012;8:e1002509.
Munshi T, Gupta A, Evangelopoulos D et al. Characterisation of
ATP-Dependent Mur Ligases Involved in the Biogenesis of
Cell Wall Peptidoglycan in Mycobacterium tuberculosis. PLoS
One 2013;8:e60143.
Nakel M, Ghuysen JM, Kandler O. Wall peptidoglycan in Aerococ-
cus viridans strains 201 Evans and ATCC 11563 and in Gaffkya
homari strain ATCC 10400. Biochemistry 1971;10:2170–5.
Nampoothiri KM, Rubex R, Patel AK et al. Molecular cloning, over-
expression and biochemical characterization of hypothetical
beta-lactamases of Mycobacterium tuberculosis H37Rv. J Appl
Microbiol 2008;105:59–67.
Ngadjeua F, Braud E, Saidjalolov S et al. Critical impact of
peptidoglycan precursor amidation on the activity of L,D-
Transpeptidases from Enterococcus faecium and M y cobac-
terium tuberculosis. Chem Eur J 2018;24:5743–7.
Nikitushkin VD, Demina GR, Shleeva MO et al. Peptidogly-
can fragments stimulate resuscitation of “non-culturable”
mycobacteria. Antonie Van Leeuwenhoek 2013;103: 37–46.
Nikolaidis I, Izoré T, Job V et al. Calcium-dependent complex
formation between PBP2 and lytic transglycosylase SltB1 of
Pseudomonas aeruginosa. Microb Drug Resist 2012;18:298–
305.
Noldeke ER, Muckenfuss LM, Niemann V et al. Structural basis
of cell wall peptidoglycan amidation by the GatD/MurT com-
plex of Staphylococcus aureus. Sci Rep 2018;8:12953.
Otten C, Brilli M, Vollmer W et al. Peptidoglycan in obligate intra-
cellular bacteria. Mol Microbiol 2018;107:142–63.
Pandey AK, Yang Y, Jiang Z et al. NOD2, RIP2 and IRF5 play a crit-
ical role in the type I interferon response to Mycobacterium
tuberculosis. PLoS Pathog 2009;5:e1000500.
Parikh A, Verma SK, Khan S et al. PknB-mediated phosphoryla-
tion of a novel substrate, N-acetylglucosamine-1-phosphate
uridyltransferase, modulates its acetyltransferase activity. J
Mol Biol 2009;386:451–64.
Park JT, Uehara T. How bacteria consume their own exoskeletons
(turnover and recycling of cell wall peptidoglycan). Microbiol
Mol Biol Rev 2008;72: 211–27, table of contents.
Park JT. Uridine-5’-pyrophosphate derivatives. II. Isolation from
Staphylococcus aureus. J Biol Chem 1952;194: 877–84.
Patru MM, Pavelka MS, Jr. A role for the class A penicillin-binding
protein PonA2 in the survival of Mycobacterium smegmatis
under conditions of nonreplication. J Bacteriol 2010;192:3043–
54.
Payen M, Muylle I, Vandenberg O et al. Meropenem-clavulanate
for drug-resistant tuberculosis: a follow-up of relapse-free
cases. Int J Tuberc Lung Dis 2018;22:34–9.
Payne DJ, Gwynn MN, Holmes DJ et al. Drugs for bad bugs: con-
fronting the challenges of antibacterial discovery. Nat Rev
Drug Discovery 2007;6:29–40.
Perdih A, Hrast M, Barreteau H et al. Benzene-1,3-dicarboxylic
acid 2,5-dimethylpyrrole derivatives as multiple inhibitors
of bacterial Mur ligases (MurC-MurF). Bioorganic Med Chem
2014;22:4124–34.
Pesnot T, Gershater MC, Ward JM et al. Phosphate mediated
biomimetic synthesis of tetrahydroisoquinoline alkaloids.
Chem Commun 2011;47:3242–4.
Petit JF, Adam A, Wietzerbin-Falszpan J et al. Chemical structure
of the cell wall of Mycobacterium smegmatis. I. Isolation and
partial characterization of the peptidoglycan. Biochem Biophys
Res Commun 1969;35:478–85.
Pink D, Moeller J, Quinn B et al. On the architecture of the gram-
negative bacterial murein sacculus. J Bacteriol 2000;182:5925–
30.
Pisabarro AG, de Pedro MA, Vazquez D. Structural modifica-
tions in the peptidoglycan of Escherichia coli associated with
changes in the state of growth of the culture. J Bacteriol
1985;161:238–42.
Piuri M, Hatfull GF. A peptidoglycan hydrolase motif within
the mycobacteriophage TM4 tape measure protein promotes
efficient infection of stationary phase cells. Mol Microbiol
2006;62:1569–85.

Plocinski P, Ziolkiewicz M, Kiran M et al. Characterization
of CrgA, a new partner of the Mycobacterium tubercu-
losis peptidoglycan polymerization complexes. J Bacteriol
2011;193:3246–56.
Poen S, Nakatani Y, Opel-Reading HK et al. Exploring the struc-
ture of glutamate racemase from Mycobacterium tuberculosis
as a template for anti-mycobacterial drug discovery. Biochem
J 2016;473:1267–80.
Pompeo F, van Heijenoort J, Mengin-Lecreulx D. Probing the
role of cysteine residues in glucosamine-1-phosphate acetyl-
transferase activity of the bifunctional GlmU protein from
Escherichia coli: site-directed mutagenesis and characteriza-
tion of the mutant enzymes. J Bacteriol 1998;180:4799–803.
Prigozhin DM, Mavrici D, Huizar JP et al. Structural and biochem-
ical analyses of Mycobacterium tuberculosis N-acetylmuramyl-
L-alanine amidase Rv3717 point to a role in peptidoglycan
fragment recycling. J Biol Chem 2013;288:31549–55.
Prosser GA, de Carvalho LP. Kinetic mechanism and inhibition of
Mycobacterium tuberculosis D-alanine:D-alanine ligase by the
antibiotic D-cycloserine. FEBS J 2013;280:1150–66.
Prosser GA, Rodenburg A, Khoury H et al. Glutamate racemase
is the primary target of β-chloro-D-alanine in Mycobacterium
tuberculosis. Antimicrob Agents Chemother 2016: AAC. 01249-16.
Quesniaux V, Fremond C, Jacobs M et al. Toll-like receptor path-
ways in the immune responses to mycobacteria. Microbes
Infect 2004;6:946–59.
Radkov AD, Hsu Y-P, Booher G et al. Imaging bacterial cell wall
biosynthesis. Annu Rev Biochem 2018;87:991–1014.
Raymond JB, Mahapatra S, Crick DC et al. Identification of the
namH gene, encoding the hydroxylase responsible for the N-
glycolylation of the mycobacterial peptidoglycan. J Biol Chem
2005;280:326–33.
Reck F, Marmor S, Fisher S et al. Inhibitors of the bacterial cell
wall biosynthesis enzyme MurC. Bioorganic Med Chem Letters
2001;11:1451–4.
Reith J, Mayer C. Peptidoglycan turnover and recycling in Gram-
positive bacteria. Appl Microbiol Biotechnol 2011;92:1–11.
Riano F, Arroyo L, Paris S et al. T cell responses to DosR and Rpf
proteins in actively and latently infected individuals from
Colombia. Tuberculosis (Edinb) 2012;92:148–59.
Rodriguez-Rivera FP, Zhou X, Theriot JA et al. Acute Modula-
tion of Mycobacterial Cell Envelope Biogenesis by Front-Line
Tuberculosis Drugs. Angew Chem Int Ed 2018;57:5267–72.
Romano M, Aryan E, Korf H et al. Potential of Mycobacterium tuber-
culosis resuscitation-promoting factors as antigens in novel
tuberculosis sub-unit vaccines. Microbes Infect 2012;14:86–95.
Romeis T, Holtje JV. Penicillin-binding protein 7/8 of Escherichia
coli is a DD-endopeptidase. Eur J Biochem 1994;224:597–604.
Roychowdhury A, Wolfert MA, Boons GJ. Synthesis and proin-
flammatory properties of muramyl tripeptides contain-
ing lysine and diaminopimelic acid moieties. ChemBioChem
2005;6:2088–97.
Ruggiero A, Marasco D, Squeglia F et al. Structure and func-
tional regulation of RipA, a mycobacterial enzyme essential
for daughter cell separation. Structure 2010;18:1184–90.
Ruiz N. Bioinformatics identification of MurJ (MviN) as the pep-
tidoglycan lipid II flippase in Escherichia coli. Proc Natl Acad
Sci 2008;105:15553–7.
Russell-Goldman E, Xu J, Wang X et al. A Mycobacterium tubercu-
losis Rpf double-knockout strain exhibits profound defects in
reactivation from chronic tuberculosis and innate immunity
phenotypes. Infect Immun 2008;76:4269–81.
Ryan NJ, Lo JH. Delamanid: first global approval. Drugs 2014;74:
1041–5.
Maitra et al. 573
Röse L, Kaufmann SH, Daugelat S. Involvement of Mycobac-
terium smegmatis undecaprenyl phosphokinase in
biofilm and smegma formation. Microbes Infect 2004;6:
965–71.
Sangshetti JN, Joshi SS, Patil RH et al. Mur Ligase Inhibitors
as Anti-bacterials: A Comprehensive Review. Curr Pharm Des
2017;23:3164–96.
Sassetti CM, Boyd DH, Rubin EJ. Genes required for mycobacte-
rial growth defined by high density mutagenesis. Mol Micro-
biol 2003;48:77–84.
Sauvage E, Kerff F, Terrak M et al. The penicillin-binding proteins:
structure and role in peptidoglycan biosynthesis. FEMS Micro-
biol Rev 2008;32:234–58.
Scheffers DJ, Pinho MG. Bacterial cell wall synthesis: new
insights from localization studies. Microbiol Mol Biol Rev
2005;69:585–607.
Scheurwater E, Reid CW, Clarke AJ. Lytic transglycosylases:
bacterial space-making autolysins. Int J Biochem Cell Biol
2008;40:586–91.
Schubert K, Sieger B, Meyer F et al. The antituberculosis drug
ethambutol selectively blocks apical growth in CMN group
bacteria. MBio 2017;8.
Schulbach MC, Brennan PJ, Crick DC. Identification of a
short (C15) chain Z-isoprenyl diphosphate synthase and
a homologous long (C50) chain isoprenyl diphosphate
synthase in Mycobacterium tuberculosis. J Biol Chem 2000:
22876-81.
Schulbach MC, Mahapatra S, Macchia M et al. Purification,
enzymatic characterization, and inhibition of the Z-farnesyl
diphosphate synthase from Mycobacterium tuberculosis. J Biol
Chem 2001;276: 11624–30.
Schulz MN, Hubbard RE. Recent progress in fragment-based lead
discovery. Curr Opin Pharmacol 2009;9:615–21.
Shah IM, Laaberki MH, Popham DL et al. A eukaryotic-like
Ser/Thr kinase signals bacteria to exit dormancy in response
to peptidoglycan fragments. Cell 2008;135:486–96.
Sham L-T, Butler EK, Lebar MD et al. MurJ is the flippase of
lipid-linked precursors for peptidoglycan biogenesis. Science
2014;345:220–2.
Shanmugam A, Anbazhagan V, Natarajan J. Virtual screen-
ing of phenylsulfonamido-3-morpholinopropan-2-yl dihy-
drogen phosphate derivatives as novel inhibitors of MurC–
MurF ligases from Mycobacterium leprae. Med Chem Res
2012;21:4341–51.
Shanmugam A, Natarajan J. Comparative modeling of UDP-N-
acetylmuramoyl-glycyl-D-glutamate-2, 6-diaminopimelate
ligase from Mycobacterium leprae and analysis of its binding
features through molecular docking studies. J Mol Model
2012;18:115–25.
Shetty A, Dick T. Mycobacterial cell wall synthesis inhibitors
cause lethal ATP burst. Front Microbiol 2018;9:1898.
Shrivastava P, Navratna V, Silla Y et al. Inhibition of Mycobac-
terium tuberculosis dihydrodipicolinate synthase by alpha-
ketopimelic acid and its other structural analogues. Sci Rep
2016;6:30827.
Siegrist MS, Whiteside S, Jewett JC et al. D-amino acid chemical
reporters reveal peptidoglycan dynamics of an intracellular
pathogen. ACS Chem Biol 2013;8: 500–5.
Siewert G, Strominger JL. Bacitracin: an inhibitor of the dephos-
phorylation of lipid pyrophosphate, an intermediate in the
biosynthesis of the peptidoglycan of bacterial cell walls. Proc
Natl Acad Sci 1967;57:767.
Silver LL. Challenges of antibacterial discovery. Clin Microbiol Rev
2011;24:71–109.
 574 FEMS Microbiology Reviews, 2019, Vol. 43, No. 5
Silver LL. Does the cell wall of bacteria remain a viable source of
targets for novel antibiotics? Biochem Pharmacol 2006;71:996–
1005.
Singh PP, Smith VL, Karakousis PC et al. Exosomes isolated from
mycobacteria-infected mice or cultured macrophages can
recruit and activate immune cells in vitro and in vivo. J
Immunol 2012;189:777–85.
Siricilla S, Mitachi K, Skorupinska-Tudek K et al. Biosynthesis of
a water-soluble lipid I analogue and a convenient assay for
translocase I. Anal Biochem 2014;461: 36–45.
Smith CA. Structure, function and dynamics in the mur family
of bacterial cell wall ligases. J Mol Biol 2006;362:640–55.
Soni V, Upadhayay S, Suryadevara P et al. Depletion of M. tubercu-
losis GlmU from Infected Murine Lungs Effects the Clearance
of the Pathogen. PLoS Pathog 2015;11:e1005235.
Stamper GF, Longenecker KL, Fry EH et al. Structure-based
optimization of MurF inhibitors. Chem Biol Drug Design
2006;67:58–65.
Strancar K, Blanot D, Gobec Set al. Design, synthesis and
structure-activity relationships of new phosphinate
inhibitors of MurD. Bioorg. Med. Chem. Lett. 2006; 16:343–
8. 16271472
Strancar K, Boniface A, Blanot D et al. Phosphinate inhibitors of
UDP-N-acetylmuramoyl-L-alanyl-D-glutamate: L-lysine lig-
ase (MurE). Arch Pharm (Weinheim) 2007;340:127–34.
Sulzenbacher G, Gal L, Peneff C et al. Crystal structure of Strepto-
coccus pneumoniae N-acetylglucosamine-1-phosphate uridyl-
transferase bound to acetyl-coenzyme A reveals a novel
active site architecture. J Biol Chem 2001;276:11844–51.
Sutcliffe IC, Harrington DJ. Lipoproteins of Mycobacterium tuber-
culosis: an abundant and functionally diverse class of cell
envelope components. FEMS Microbiol Rev 2004;28:645–59.
Taguchi A, Welsh MA, Marmont LS et al. FtsW is a peptidoglycan
polymerase that is functional only in complex with its cog-
nate penicillin-binding protein. Nature Microbiol 2019:4:587–
94.
Takayama K, Schnoes HK, Semmler EJ. Characterization of the
alkali-stable mannophospholipids of Mycobacterium smegma-
tis. Biochem Biophys Acta 1973;316:212–21.
Talukder AA, Yanai S, Nitta T et al. RpoS-dependent regulation
of genes expressed at late stationary phase in Escherichia coli.
FEBS Lett 1996;386:177–80.
Tanner ME, Vaganay S, van Heijenoort J et al. Phosphinate
inhibitors of the D-glutamic acid-adding enzyme of peptido-
glycan biosynthesis. J Org Chem 1996;61:1756–60.
Templin MF, Ursinus A, Holtje JV. A defect in cell wall recy-
cling triggers autolysis during the stationary growth phase
of Escherichia coli. EMBO J 1999;18:4108–17.
Thanky NR, Young DB, Robertson BD. Unusual features of
the cell cycle in mycobacteria: polar-restricted growth and
the snapping-model of cell division. Tuberculosis (Edinb)
2007;87:231–6.
Tomasic T, Zidar N, Kovac A et al. 5-Benzylidenethiazolidin-4-
ones as multitarget inhibitors of bacterial Mur ligases. Chem
Med Chem 2010;5:286–95.
Tomašić T, Mašič LP. Rhodanine as a scaffold in drug discovery: a
critical review of its biological activities and mechanisms of
target modulation. Expert Opin Drug Discovery 2012;7:549–60.
Triboulet S, Bougault CM, Laguri C et al. Acyl acceptor recognition
by Enterococcus faecium L,D-transpeptidase Ldtfm. Mol Micro-
biol 2015;98:90–100.
Trott O, Olson AJ. AutoDock Vina: improving the speed and accu-
racy of docking with a new scoring function, efficient opti-
mization, and multithreading. J Comput Chem 2010;31:455–61.
Turapov O, Forti F, Kadhim B et al. Two Faces of CwlM, an Essen-
tial PknB Substrate, in Mycobacterium tuberculosis. Cell reports
2018;25:57–67 e5.
Turk S, Kovac A, Boniface A et al. Discovery of new inhibitors of
the bacterial peptidoglycan biosynthesis enzymes MurD and
MurF by structure-based virtual screening. Bioorganic Med
Chem 2009;17:1884–9.
Turner RD, Mesnage S, Hobbs JK et al. Molecular imaging of gly-
can chains couples cell-wall polysaccharide architecture to
bacterial cell morphology. Nature Commun 2018;9:1263.
Udou T, Ogawa M, Mizuguchi Y. Spheroplast formation of
Mycobacterium smegmatis and morphological aspects of their
reversion to the bacillary form. J Bacteriol 1982;151:1035–9.
Uehara A, Fujimoto Y, Kawasaki A et al. Meso-diaminopimelic
acid and meso-lanthionine, amino acids specific to bacte-
rial peptidoglycans, activate human epithelial cells through
NOD1. J Immunol 2006;177:1796–804.
Usha V, Dover LG, Roper DI et al. Structure of the diaminopime-
late epimerase DapF from Mycobacterium tuberculosis. Acta
Crystallogr Sec D 2009;65:383–7.
Usha V, Lloyd AJ, Lovering AL et al. Structure and function of
Mycobacterium tuberculosis meso-diaminopimelic acid (DAP)
biosynthetic enzymes. FEMS Microbiol Lett 2012;330:10–6.
Usha V, Lloyd AJ, Roper DI et al. Reconstruction of
diaminopimelic acid biosynthesis allows characterisation of
Mycobacterium tuberculosis N-succinyl-L, L-diaminopimelic
acid desuccinylase. Sci Rep 2016;6:23191.
Vaganay S, Tanner ME, van Heijenoort J et al. Study of the reac-
tion mechanism of the D-glutamic acid-adding enzyme from
Escherichia coli. Microb Drug Resist 1996;2:51–4.
van Asselt EJ, Thunnissen AM, Dijkstra BW. High resolution
crystal structures of the Escherichia coli lytic transglycosylase
Slt70 and its complex with a peptidoglycan fragment. J Mol
Biol 1999;291:877–98.
van Heijenoort J. Formation of the glycan chains in the synthesis
of bacterial peptidoglycan. Glycobiology 2001a;11:25R–36R.
van Heijenoort J. Lipid intermediates in the biosynthesis of bac-
terial peptidoglycan. Microbiol Mol Biol Rev 2007;71:620–35.
van Heijenoort J. Recent advances in the formation of the bacte-
rial peptidoglycan monomer unit. Nat Prod Rep 2001b;18:503–
19.
Vicente M, Rico AI, Martinez-Arteaga R et al. Septum enlight-
enment: assembly of bacterial division proteins. J Bacteriol
2006;188:19–27.
Vijayrajratnam S, Pushkaran AC, Balakrishnan A et al. Bacte-
rial peptidoglycan with amidated meso-diaminopimelic acid
evades NOD1 recognition: an insight on NOD1 structure-
recognition study. Biochem J 2016:473:4573–92.
Vincent AT, Nyongesa S, Morneau I et al. The mycobacterial cell
envelope: a relict from the past or the result of recent evolu-
tion? Front Microbiol 2018;9:2341.
Volkamer A, Eid S, Turk S et al. Pocketome of human kinases:
prioritizing the ATP binding sites of (yet) untapped protein
kinases for drug discovery. J Chem Inf Model 2015;55:538–49.
Vollmer W, Holtje JV. The architecture of the murein (peptidogly-
can) in gram-negative bacteria: vertical scaffold or horizontal
layer(s)? J Bacteriol 2004;186:5978–87.
Vollmer W, Joris B, Charlier P et al. Bacterial peptidoglycan
(murein) hydrolases. FEMS Microbiol Rev 2008;32:259–86.
Vollmer W, von Rechenberg M, Holtje JV. Demonstration of
molecular interactions between the murein polymerase
PBP1B, the lytic transglycosylase MltA, and the scaffold-
ing protein MipA of Escherichia coli. J Biol Chem 1999;274:
6726–34.

Vollmer W. Structural variation in the glycan strands of bacterial
peptidoglycan. FEMS Microbiol Rev 2008;32:287–306.
von Rechenberg M, Ursinus A, Holtje J-V. Affinity chromatogra-
phy as a means to study multienzyme complexes involved
in murein synthesis. Microb Drug Resist 1996;2:155–7.
Votyakova TV, Kaprelyants AS, Kell DB. Influence of viable cells
on the resuscitation of dormant cells in micrococcus luteus cul-
tures held in an extended stationary phase: the population
effect. Appl Environ Microbiol 1994;60:3284–91.
Wang Q, Matsuo Y, Pradipta AR et al. Synthesis of characteristic
Mycobacterium peptidoglycan (PGN) fragments utilizing with
chemoenzymatic preparation of meso-diaminopimelic acid
(DAP), and their modulation of innate immune responses.
Organic Biomol Chem 2016;14:1013–23.
Wang W, Dong C, McNeil M et al. The structural basis of chain
length control in Rv1086. J Mol Biol 2008;381:129–40.
WHO. Global Tuberculosis Report Geneva. Geneva, Switzerland:
World Health Organisation, 2018.
Wietzerbin J, Das BC, Petit JF et al. Occurrence of D-alanyl-
(D)-meso-diaminopimelic acid and meso-diaminopimelyl-
meso-diaminopimelic acid interpeptide linkages in the pep-
tidoglycan of Mycobacteria. Biochemistry 1974;13:3471–6.
Wolfert MA, Roychowdhury A, Boons G-J. Modification of the
structure of peptidoglycan is a strategy to avoid detection by
nucleotide-binding oligomerization domain protein 1. Infect
Immun 2007;75:706–13.
Workman SD, Worrall LJ, Strynadka NC. Crystal structure of an
intramembranal phosphatase central to bacterial cell-wall
peptidoglycan biosynthesis and lipid recycling. Nature Com-
mun 2018;9:1159.
Wube AA, Hüfner A, Thomaschitz C et al. Design, synthesis and
antimycobacterial activities of 1-methyl-2-alkenyl-4 (1H)-
quinolones. Bioorganic Med Chem 2011;19:567–79.
Xie J, Hodgkinson JW, Katzenback BA et al. Characterization
of three Nod-like receptors and their role in antimicrobial
responses of goldfish (Carassius auratus L.) macrophages to
Aeromonas salmonicida and Mycobacterium marinum. Dev Comp
Immunol 2013;39:180–7.
Xu L, Wu D, Liu L et al. Characterization of mycobacterial UDP-
N-acetylglucosamine enolpyruvyle transferase (MurA). Res
Microbiol 2014;165:91–101.
Maitra et al. 575
Yang Y, Severin A, Chopra Ret al. 3,5-dioxopyrazolidines, novel
inhibitors of UDP-N- acetylenolpyruvylglucosamine reduc-
tase (MurB) with activity against gram-positive bacteria.
Antimicrob. Agents Chemother 2006;50:556–64. 16436710
Yan Y, Munshi S, Leiting B et al. Crystal structure of Escherichia coli
UDPMurNAc-tripeptide d-alanyl-d-alanine-adding enzyme
(MurF) at 2.3 A resolution. J Mol Biol 2000;304:435–45.
Young DB, Perkins MD, Duncan K et al. Confronting the scien-
tific obstacles to global control of tuberculosis. J Clin Invest
2008;118:1255–65.
Zapun A, Philippe J, Abrahams KA et al. In vitro reconstitution
of peptidoglycan assembly from the Gram-positive pathogen
Streptococcus pneumoniae. ACS Chem Biol 2013;8:2688–96.
Zawadzke LE, Bugg TD, Walsh CT. Existence of two D-alanine:D-
alanine ligases in Escherichia coli: cloning and sequencing
of the ddlA gene and purification and characterization of the
DdlA and DdlB enzymes. Biochemistry 1991;30:1673–82.
Zawadzke LE, Norcia M, Desbonnet CR et al. Identification of an
inhibitor of the MurC enzyme, which catalyzes an essen-
tial step in the peptidoglycan precursor synthesis pathway.
Assay Drug Dev Technol 2008;6: 95–103.
Zhang Y, Heym B, Allen B et al. The catalase-peroxidase gene
and isoniazid resistance of Mycobacterium tuberculosis. Nature
1992;358:591–3.
Zhang Y, Yang Y, Woods A et al. Resuscitation of dormant
Mycobacterium tuberculosis by phospholipids or specific pep-
tides. Biochem Biophys Res Commun 2001;284:542–7.
Zhang Y. The magic bullets and tuberculosis drug targets. Annu
Rev Pharmacol Toxicol 2005;45:529–64.
Zhang Z, Bulloch EM, Bunker RD et al. Structure and function of
GlmU from Mycobacterium tuberculosis. Acta crystallogr Sec D,
Biol Crystallogr 2009;65:275–83.
Zheng S, Sham LT, Rubino FA et al. Structure and mutagenic anal-
ysis of the lipid II flippase MurJ from Escherichia coli. Proc Natl
Acad Sci U S A 2018;115:6709–14.
Zidar N, Tomasic T, Sink R et al. Discovery of novel 5-
benzylidenerhodanine and 5-benzylidenethiazolidine-2,4-
dione inhibitors of MurD ligase. J Med Chem 2010;53:6584–94.
Zumla A, Nahid P, Cole ST. Advances in the development of new
tuberculosis drugs and treatment regimens. Nat Rev Drug Dis-
covery 2013;12:388–404.